Dell PowerConnect 7024P, PowerConnect M6348, PowerConnect 7048R, PowerConnect 7024F, PowerConnect 8024F User Manual

...

Dell™ PowerConnect™ 8132/8164/8132F/8164F/7024/7048/7024P/7048P/7024F/7048R/ 7048R-RA/8024/8024F/M6220/M6348/M8024/M8024-k

PowerConnect

5.1.1.7 Firmware Release Notes

Date: July 2013 System Firmware Version 5.1.1.7

Reproduction in any manner whatsoever without the written pe r m is sion of Dell I nc. is strictly forbidden.

Trademark s used in this t ex t: Dell, the DE LL logo and PowerC o nnect are trademarks of D el l Inc; Intel and Pentium ar e r eg istered trademarks and Celeron is a trademark of Intel Corporation; Microsoft and Windows are registered trademarks of Microsoft Corporation. Other trademarks and tr ade names may be used in this document to refer to either the entity claiming the marks and names or thei r products. Dell

Inc. disclaims any proprietary interest in trademarks and trade names other than its own. All rights reserved. This document m ay not, in whole or in part, be copied, photocopied, reproduced, translated, or reduced to any electronic medium or machine-readable form without the prior written consent of Dell. Dell reserves the right to make changes without further notice to any products or specifications referred to herein to im prove reliability, functionality or design.

Reproduction, adaptation or translation without prior written permission is prohibited, except as allowed under the copyrigh t laws.

Introduction 1 Global Support 1 Firmware Specifications 1 Firmware Upgrade 2 Firmware Downgrade 3 Boot Code Downgrade 4 Hardware Supported 4 Support Matrix 5 Supported Firmware Functionality 6 Added Functionality in this Release 7 Changed Functionality in this Release 14 Issues Resolved 22 CLI Reference Guide Updates 55 User’s Configuration Guide Updates 59 Known Issues 61 Known Restrictions and Limitations 70 System – 5.0.1.3 71 System – 5.0.0.4 71 Management – 4.2.1.3 72 Layer 2 – 4.2.0.4 73 Layer 3 – 4.2.0.4 74 Management – 4.2.0.4 74 Data Center – 4.2.0.4 75 End of Release Notes 76

2 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

PC8100v5.1.1.7.stk

PCM8024kv5.1.1.7.stk

5.1.1.7

July 2013

Version Numbering Convention

Version number

Description

PowerConnect

5 1 1

Four part version number



Denotes an ad hoc release of the product software.

Denotes a scheduled maintenance release of the product software.

Denotes a major version number.

Introduction

This document provides specific information for the Dell PowerConnect 8132/8164/8132F/8164F/7024/7048/7024P/7048P/7024F/7048R/7048R-RA/8024/8024F/M6220/M6348/M8024/M8024-k switches firmware versi on 5.1.1.7.

It is recommended that this release note be thoroughly reviewed prior to installing or upgrading of this product.

Global Support

For information regardi ng the latest available firmware, release note revisio ns, or additional assistance, please visit

support.dell.com.

Firmware Specifications

Firmware Version

Firmware Image Name Version Number Release Date

PCM6220v5.1.1.7.stk

PC7000_M6348v5.1.1.7.stk

PC8024v5.1.1.7.stk

PCM8024v5.1.1.7.stk

Series



5.1.1.7

Denotes the build number.

July 2013 July 2013 July 2013 July 2013

System Firmware Version 5.1.1.7 Page 1

Firmware Upgrade

NOTE: Administrators upgrading PowerConnect 7024/7048/7024P/7048P/7024F/

7048R/7048R-RA/8024/8024F/M6220/M6348/M8024/ M8024k switches from 2.x.x.x or

3.x.x.x or 4.x.x.x or 5.x.x.x versions of firmware MUST follow the instructions documented in the

Upgrading PowerConnect Switches from Version 2.x.x.x or 3.x.x.x

or 4.x.x.x or 5.x.x.x to 5.1.1.7 Firmware procedure. Failure to follow the procedure s

described in that document when upgrading from 2.x.x.x or 3.x.x.x or 4.x. x.x or 5.x.x.x firmware may result in an inoperable switch!

NOTE: After upgradi ng the switch firmware version to 5.1.1.7, the CPLD (Complex

Programmable Logic Device) code update is required on M8024-k and P C8024/P C80 24F switches via serial console if switch is running with the older CPLD version. The latest CPLD version available for M8024k switch is Version 5 and for PC8024/PC8024F switch is Version 6. Administrators upgrading PowerConnect 8024/ 8 0 2 4F/M8024k switches MUST follow the CPLD update instructions documented in the

Upgrading PowerConnect

Switches from Version 2.x.x.x or 3.x.x.x or 4.x.x.x or 5.x.x.x to 5.1.1.7 Firmware

procedure.

NOTE: Administrators upgrading PowerConnect 8132/8164/8132F/8164F switches

MUST follow the instr u ctions documented in the

Upgrading PowerConnect

8132_8164_8132F_8164F switche s from Version 5.x.x.x to 5.1.1.7 Firmware procedure. Failure t o follow the pro cedures described in that document w h en upgrading firmware may result in an inoperable switch!

NOTE: OMNM (Open Manage Network Manager) v5.2 SP1 supports firmware

management of PowerConnect 7024/7048/7024P/7048P/7024F/7048R/7048RRA/8024/8024F/M6220/M6348/M8024/M8024-k to deploy the firmware version 4.1.x.x or later. OMNM v5.2 SP1 supports PowerConnect 8132/8164/8132F/8164F t o deploy the firmware version 5.0.0.4

2 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Part Numbers for PowerConnect 81xx

A1 CPU

B1 CPU

PowerConnect 8132 (US, Canada, Mexico, South Ame r ic a)

TRJ78

PTM0F

PowerConnect 8132F

W0HV1

NWHGV

PowerConnect 8132F (all other countries)

7D1GN

KWHG3

H0F6C

N00C1

PowerConnect 8164 (all other countries)

P8RHX

Y2FJ0

PowerConnect 8164F (US, Canada, Mexico, South America)

VTWN8

4PHP2

PowerConnect 8164F

1JWM5

8KHT1

Firmware Downgrade

Downgrading from 5.1.1.7 to an earlier release is supported on most PowerConnect series switches (except PC81xx series with B1 CPU versions, see note below); however, migration of configuration information from a later release to an earl ier

release is not supported. The existing configuration may or may not work with the earlier version of firmware, therefore, it is best to be physically present at the switch site and to be prep ar ed to access the switch over the s er ial port if necessary when downgrading firmware.

Auto-downgrade of a stack is not enabled by default. If downgrading a stack, be sure to enable auto-downgrade before activating the earlier versions of firmware on the stack master.

Recent versions of the PowerConnect 81xx series switches support newer versions of CPU (B1). The B1 version of CPU requires firmware release version 5.1.0.1 or later and cannot be downgraded to earlier firmware releases. In addition, if this unit is to be deployed as a member within a stack, the entire stack will be required to run 5.1.0.1 or later firmware. Here are some st eps to help determ i ne what CPU is in the PowerConnect 8100 Series switch:

1. Run show version from the CLI prompt.

2. If there is no CPU version line, then t he stack is r unning 5.0.x.x firmware . Since 5.0.x.x can onl y run on

A1 CPU switches, then all switches in the stack (or a standalone) contain A1 CPUs.

3. If there is a CPU version line (like below):

CPU Version....................... XLP308H-A1

(It will display the CPU version in the last two characters - either A1 or B1).

4. On a stack, r un show version m where m is the unit number assigned to the stack member to see each CPU

version for each member in the stack.

In addition here are the part numbers for reference:

PowerConnect 8132 (all other countries) 0C90P X20W5

(US, Canada, Mexico, So u th America)

PowerConnect 8164 (US, Canada, Mexico, South Ame r ic a)

(all other countries)

Recent versions of the PowerConnect M8024-K modular switches have hardware changes that require firmware version 4.1.0.19 or higher. Power C onne c t M 8 02 4-K with a new PPID label cannot be downgraded to the firmware version below 4.1.0.19. If a downgrade is attempted, the firmware activatio n procedure will detect th at earlier firmware version is not compatible with the switch hardware and abort the activation procedure. Any PowerConnect M8024-K with the old PPID label can accept any version of the switch firmware.

MODEL OLD PPID NEW PPID

M8024K

System Firmware Version 5.1.1.7 Page 3

2F07F A00 2F07F A01 57821 A00 57821 A01

Boot Code Downgrade

Never downgrade the boot code! The 5.1.1.7 boot code supports all earlier versions of firmware and never needs downgrading.

Hardware Supported

• Dell PowerConnect 8132 Ethernet Switch

• Dell PowerConnect 8164 Ethernet Switch

• Dell PowerConnect 8132F Ethernet Switch

• Dell PowerConnect 8164F Ethernet Switch

• Dell PowerConnect M6220 Ethernet Switch

• Dell PowerConnect M6348 Ethernet Switch

• Dell PowerConnect 7024 Ethernet Switch

• Dell PowerConnect 7048 Ethernet Switch

• Dell PowerConnect 7024P Ethernet Switch

• Dell PowerConnect 7048P Ethernet Switch

• Dell PowerConnect 7024F Ethernet Switch

• Dell PowerConnect 7048R Ethernet Switch

• Dell PowerConnect 7048R-RA Ethernet Switch

• Dell PowerConnect 8024 Ethernet Switch

• Dell PowerConnect 8024F Ethernet Switch

• Dell PowerConnect M8024 Ethernet Switch

• Dell PowerConnect M8024-k Ethernet Switch

4 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

  



  



  



  



  



  

Support Matrix

Since not all functionality is supported on all switches, the following matrix identifies the major differences among the PowerConnect switch models. A check mark indicates support for the feature. All other features listed in the release notes ar e supported on all switches.

Feature/Switch

Dell PowerConnect M6220 Ethernet Switch

Dell PowerConnect M6348 Ethernet Switch

Dell PowerConnect 7024 Ethernet Switch

Dell PowerConnect 7048 Ethernet Switch

Dell PowerConnect 7024P Ethernet Switch

Dell PowerConnect 7048P Ethernet Switch

Dell PowerConnect 7024F Ethernet Switch

Dell PowerConnect 7048R Ethernet Switch

Dell PowerConnect 7048R-RA Ethernet Switch

Dell PowerConnect 8024 Ethernet Switch

Dell PowerConnect 8024F Ethernet Switch

Dell PowerConnect M8024 Ethernet Switch

Dell PowerConnect M8024-k Ethernet Switch

Dell PowerConnect 8132 Ethernet Switch

Dell PowerConnect 8164 Ethernet Switch

Dell PowerConnect 8132F/8164F Ethernet Switch

Priority

Flow

Control

DCBx ETS PoE+

iSCSI

Optimization

 

  



 



   

  



USB

      

  

grEEEn

Ethernet

Hot

Swap

Cards

WRED



   

System Firmware Version 5.1.1.7 Page 5

Supported Firmware Functionality

For more details regarding all the supported firmware features and functionality, please refer to the Dell PowerConnect Series CLI Reference Guide and the Dell PowerConnect Series User’s Configuration Guide.

6 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Added Functionality in this Release

This section contains a list of features added in this rele ase that are new for at least one switch listed in the Hardware Supported section above.

Release 5.1.1.7

 No Added Functionality in this Release

Release 5.1.0.1

 IGMP Snooping Improvements  IP Multicast Enhancements  Support for B1 CPU – only on PC8100 series  New Browser Support

• Mozilla Firefox 14

• Internet Explorer 9

• Google Chrome 21

 Static Route Maximum Increased to 512 (for all switches except for M6220 series)  USB Auto-configuration expanded for multiple MAC Address support  Ability to reset stack port counters  Increased Maximum number of dot1x clients per port to 24  Commands to remove signed certificates/Keys  Support for Additional transceivers/optics

Release 5.0.1.3

 Added PoE DC Disconnect HW support – on PC7000 series

Release 5.0.0.4

 Added Native EEE Support for the PC8100 10GBaseT Ports  802.1Qaz (ETS) – on 8100 series  Support for 40 Gig (QSFP+) interfaces  Support for Bootcode upgrade from pre-4.x image to 5.x image without manual system reset.  Local Preference for LAG  Private VLAN  CLI output filtering  Routing Improvements for OSPF  UDLD  Administrative Profiles  AAA Authorization  TACACS+ Ac countin g  Stacking over QSFP+ ports  QSFP+ diagnostics  sFlow Support on Port Channels

System Firmware Version 5.1.1.7 Page 7

Release 4.2.2.3

 No Added Functionality in this Release

Release 4.2.1.3

 No Added Functionality in this Release

Release 4.2.0.4

 DCBx on M8024-k/8024/8024F  FIP Snooping Bridge on M8024-k/8024/8024F  Add EEE Support for the PC7000 10GBaseT Ports  GUI EEE Power Savings Charts  RP Failover Performance Improvement  Flexible Dynamic LAG Limits  CLI Help Usability Improvements  Ability to Show Static Route Entries  CMC XML Support  Stacking Over Ethernet Por ts  Change Dell EqualLogic iSCSI Auto Detect to default  Add CLI Macro (Port Profile) for Dell Compellent Storage Equipment  Provide Dell EqualLogic DCBx TLV Auto Detect and Configuration on M8024-k/8024/8024F

Release 4.1.1.9

 No Added Functionality in this Release

Release 4.1.0.19

 Media-type CLI command

Added new command "media-type" to co nfigure an interface to select the specified media on a combo port. It is recommended the administrators select the specific media type for the particular type of network connection they expect to use. Users may observe a single port flap when the media type is changed from RJ45 to auto-select RJ45 and only the corresponding SFP port is enabled. Refer to the CLI Reference Guide Up dates section below for th e complete syntax.

Release 4.1.0.6

 IPv4-Only Mode Optimization

PowerConnect switches allocate the maximum sizes for routing tables (and others, as applicable) for both IPv4 and IPv6. Switch Performance Optimization allows the operator to optimize the allocation of switch silicon tables for either IPv4 only or mixed IPv4/IPv6 operation. The template specified limits are enforced by routing components when routes are bein g l earned. When IPv4 only mode is select ed, the following capabilities are disabled:

• DHCPv6 r elay

• DHCPv6 server

• IPv6 routing/forwarding

• OSPFv3

• IPv6 Neighbor Discovery

• Configured v6-over-v4 tunnels

• Automatic (6to4) tunnels

8 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Blocked Destination

MAC Address

Blockcdp

Used to block CDP PDU’s

N/A

0x2000

Blockvtp

Used to block VTP PDU;s

N/A

0x2003

Blockdtp

Used to block DTP PDU’s

N/A

0x2004

Blockudld

Used to block UDLD PDU’s

N/A

0x0111

Blockpagp

Used to block PAGP PDU’s

N/A

0x0104

Blocksstp

Used to block SSTP PDU’s

N/A

0x010b

Blockall

Used to block all defined Protocol Filtering PDU’s

01:00.0C:CC:CC:C0

N/A

 Auto-Install

 Link Local Protocol Filtering

• IPv6 Multicast

A reboot is required when changing to or from IPv4 mode.

USB based auto-install is an easy way to quickly bring up a switch with a known configuration. Network based autoinstall is useful in rolling out a configuration or firmware update to a group of switches or in maintaining a central repository of switch configurations and firmware where the switches always obtain their firmware and configuration from a central server.

The following clarifications are helpful in understanding the processing steps in auto-install:

• Always power on the switch that is desired to be the stack master first

• Auto-install never p ro ceeds if a startup-config file is present o n the (master) switch

• USB auto -install is attempted first. Network auto-install only proceeds if USB auto-install fails.

• If there are multiple .setup files present on the USB flash device, the powerconnect.setu p file is selected

• If a valid .setup file is not found on the USB flash device, the single .text file is used

• If multiple .text files are present, the powerconnect.text file is used.

Network based auto-install utilizes information obtained from a DHCP server. Refer to the documentation for a discussion of the DHCP options used by Auto-Install.

When auto-install downloads a firmware image to switch memory, it compares the version to the current switch image. If different, the image in memory is copied to the switch backup image and activatio n of the image is attempted. If activation succeeds, the switch is rebooted and auto-install then attempts configuration file download.

Auto-in s tall configuration files are executed as a script. For more details on Auto-Install, refer to the User’s Guide.

Link Local Protocol Filtering blocks Cisco link local protocols from being flooded in the network. By default, PowerConnect switches process and r es pond to Cisco CDP packets. However, i n networks where this capability is not desirable or other Cisco proprietary packets are flooded over the network, the administrator can disable flooding of Cisco link local protocols. The following table identifies the matching criteria for filtering Cisco proprietary packets:

Rule Type Rule Purpose

System Firmware Version 5.1.1.7 Page 9

Ether Type

 DHCP Server

The PowerConnect Series switches support a si mpl e DH C P server capability for domains that do not wish to deploy a redundant DHCP address assignment solution or who have need of a temporary solution while (re)deploying their DHCP server solution.

In configuring DHCP scopes, be aware that the DHCP pool address and netmask must exactly match a VLAN address and netmask assignment for D H CP addresses to be served over that VLAN.

Only a single manual IP address can be assigned to a pool. The address must have a netmask of 32 .

 GMRP

The GARP Multicast Registration Protocol provides a mechanism that allows networking devices to dynamically register (and de-register) Group membership information with the MAC networking devices attached to the same segment, and for that information to be disseminated across all networking devices in the bridged LAN that support Extended Filtering Services. The PowerConnect Series switches support GMRP as specified in IEEE 802.1Q 1998.

 WRED

Weighted Random Early Drop is supported on certain PowerConnect series switches. Refer to the table at t he beginning of this section for further information. CoS queue configuration involves the following hardware port queue configuration parameters:

• schedul er type: strict vs. weighted

• minimum guaran teed bandwidth

• maximum allowed bandwidth (i.e. shaping)

• queue manage ment type: tail drop vs. WRED

• tail drop par ameters: threshold

• WRED parameters: minimum threshold, maximum threshold, drop probability

Tail drop and WRED parameters are specified individually for each supported drop precedence level. In addition, the following are specified on a per-interface basis:

• queue management type: tail drop vs. WRED (only if per-queue configuration is not supported)

• WRED decay exponent

Switch administrators should remember to configure ingress ports as trusted or un-trusted. By default ingress ports trust dot1p values.

 Stack Firmware Synchronization

Stack firmware synchronization updates all stack members to the active firmware versi on on the master switch. Stack firmware synchronization is enabled b y defaul t. Stack firmware downgrade is enabled by default.

 Multicast VLAN Registration

Multicast VLAN Registration provides a method of coalescing multicast traffic requested by users on multiple VLANs onto a single VLAN when carried over the network.

MVR does not require t ha t ei t her source or recei v e r ports uti li z e V LAN tagging.

Network planners are re min ded that multicast groups in the 224.0.0.x range are reserved for multicast control plane traffic. Network planners should select multicast groups in another range for normal multicast traffic, e.g. 239.0.1.x

 iSCSI Optimization

iSCSI Optimization automatically configures ports for use with the iSCSI protocol and tracks iSCSI sessions on the PowerConnect 7000 and 8000 Series switches as well as t he PCM6348. Dell EqualLogic arrays are automatically detected and configuration of Dell EqualLogic connected ports is perfor med automatically.

Administrators are advised that the configuration performed by enabling iSCSI optimization is not automatically reversed on disabling the feature. The administrator will need to manually remove the configurat ion s e t ti ng s w he n migrating Dell EqualLogic servers or iSCSI initiator ports to other ports or switches.

10 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Detection of Dell EqualLogic arrays is keyed on receipt of the mandatory System Description TLV in the LLDP packet. Disabling LLDP will effectively disable Dell EqualLogic array detection.

Dell EqualLogic arra ys are required to be upgraded to fir mwar e 5.0.2 or later in order to use the iSCSI Optimization feature.

 LLDP

Administrators should ensure that LLDP-MED is enabled in order to operate EEE. Disabli ng LLDP or LLD P-MED will effectively disable EEE, IEEE 802.3at PoE+ high power negotiation and Dell EqualLogic array detection in the iSCSI Optimization feature.

 Connectivity Fault Management

Connectivity Fault Management performs Metro Et hernet maintenance functions. Dell PowerConnect CFM supports the following functions defined in IEEE 802.1ag Draft 8.1:

• Path disco very (link trace messages)

• Fault detect ion (continuity check messa ge)

• Fault verification and isolation (loopback and link trace messages)

• Fault notification (alarm indication signal or SNMP trap).

 Management IP Address Conflict Detection

Management IP address conflict detection activel y lo oks for duplicate IP address as s ignment and logs conflicts. Only the last identified IPv4 address conflict is retained for display by a show command. Administrators may examine the in- memory logs or the output from a SYSLOG server to identify the historical IP address conflicts. If console logging is enabled for traps, a message will appear on the console indicating that an address conflict has o ccurred.

 Email Alerting

Email alerting allows administrators to be notified via email regarding system events. Multiple email addresses can be configured. The system will attempt to resol ve mai l servers specified with a FQDN immediately and, if su ccess fu l, store the mail-server as an I P address. If a new IP address is subsequently assigned t o the mail server, the oper ator will need to re-assign the email address on the switch.

Only the Mail User Agent functionality of RFC 4409 is implemented. The PowerConnect switch does not implem e nt SMTP server functionality.

 802.1X Monitor Mode

Monitor mode is a special debug mode that assists network administrators in configuring 802.1X authenticators. Users attempting to authenticate using the authenticator are always granted access when monitor mode is enabled. All interactions with th e s upplicant and the authenti cation server are logged.

Administrators are cautioned against enabling monitor mode in a deployed network where 802.1X users may gain access to sensitive n etwork resources.

 Time Controlled ACLs

Time controlled ACLs allow administrators to apply ACLs based on the time of day. Both periodic and absolute time periods may be configured.

Administrators are cautioned that invalid (overlapping) periodic entries within a time range will prevent the time range from being applied. Administrators are advised to t es t their periodic entries an d validate that they become active as expected before deploying the time ranges in a production network. Administrators can check if a time range is act ive by using the show time-range command.

It is recommended to enable ACL logging to ensure notice of ACL activation and de-activation.

System Firmware Version 5.1.1.7 Page 11

 SNTP over IPv6

SNTP operates over IPv4 and IPv6 and may be configured using IPv4 or IPv6 addresses or DNS.

 Strong Passwords

The strong passwords featu r e allows administrators to specify that local switch pas swords meet certain characteristics considered to enhance network security.

Administrators are advi sed that the minimum character classes configuration must be enabled (value equal to 1 or greater) along with enabling the strong password feature before the other minimum character class configurations are enforced. These character class configurations are:

• Minimum number of uppercase letters.

• Minimum number of lowercase letters.

• Minimum number of numeric characters.

• Minimum number of special characters

The password strength restrictions do not apply to users configured for the internal authentication server.

 Switch Auditing

Switch auditing enhances network security by logging sensitive administrative actions. Switch auditing logs the following actions:

• Successful login

• Unsuccessful attempt to login

• Logout out from the switch

• Timed out logout from the switch

• Download file to the switch

• Upload file from the switch

• Remove file from the flash

• File changes on the flash

• Clear configu ration

• Add or remove user

• Change user acces s level

Use of a SYSLOG server for monitoring network events is highly recommended.

 Authentication

The PowerConnect switches support authentication via a number of methods. The methods are specified in named lists. Lists may be assigned to the enable and login access methods. The supported authentication methods are:

• Enable

• Line

• RADIUS

• TACACS

• IAS

• Local

• None

Methods are attempted in the order specified in the authentication list. If the authentication method rejects authentication, the user login is rejected. If an authentication method fail s , e.g. unable to contact the authentication server, the next method in the list is attempted. The IAS, local and none methods can never fail so, if specified, must be last in the list.

The 802.1X authentication list cannot be named and only supports the RADIUS, IAS, or none authentication methods. The 802.1X authentication can only have a single method.

12 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

 Internal Authentication Server

The PowerConnect Series switches support 802.1X authentication of network users from an internal authentication database. IAS users are given access to network resources. IAS is not a valid method for login or enable authentication.

The IAS database can be downloaded to the switch using the “ias-users” target in the copy command. The i as -users file takes the form of a configuration script, as follows:

configure aaa ias-user username client-1 password my-password1 exit aaa ias-user username client-2 password aa5c6c251fe374d5e306c62496c3bcf6 encrypted exit aaa ias-user username 1f3ccb1157 password 1f3ccb1157 exit

IAS users may also be configured via the web interface.

 DNS Client

The PowerConnect Series switches support name resolu tion via an embedded DNS client. When a DNS name is specified, it is attempted to be resolved against the configured DNS servers immediately. The PowerConnect switches will store the resolved IP address. If the IP address of the host resolved via DNS changes, th e administrator will need to update the configured IP address, either via DNS or manually.

If the switch is configured to obtain an address via DHCP, DNS server information received from the DHCP server is used to populate the DNS client configuration.

 Port Profiles (CLI Macros)

The PowerConnect series of switches provid es a convenient way to save and share common configurations through the use of CLI macros. A CLI macro is a set of commands having a unique name. When a CLI macro is applied, the CLI commands contained within the macro are executed and added to the running configuration. When the macro is applied to an interface, th e exi sting interface configur ations are not lost; the new commands are added to the interface and are saved in the running configuration.

A CLI macro may have keywords (variables) which are replaced by values provided when th e macro is applied (up to 3 keywords per macro). M acros can be applied to specific i nterfaces, a range of int er faces, or the global configuration.

Administrators may add their own macros or utilize the built-in macros. Administrators are cautioned to ensure that a macro does not change command modes (e.g., change from inter fa ce

configuration mode to global configuration mode). The software includes 6 built-in macros:

• profile-global - the global configuration used to enable RSTP and loop guard.

• profile-desktop - the int er face co nfiguration for incr eased network security and reliability when connecting a

desktop device, such as a PC, to a switch port.

• profile-phone - the in terface configuration used when connecting a desktop device such as a PC with an IP

phone to a switch port.

• profile-switch - the interface confi guration used when connecting an access switch and a distribution switch

or between access switches.

• profile-router - the interface configur ation used when connecting the switch and a WAN router.

• profile-wireless- the interface configur ation used when connecting the switch and a wireless access point.

Built-in macros may not be deleted or altered by the operator.

System Firmware Version 5.1.1.7 Page 13

Changed Functionality in this Release

This section contains commentary on significant differences from previous releases of firmware on PowerConnect switches, e.g. the 8132/8164/8132F/8164F/7024/7048/7024P/7048P/7024F/7048R/7048R-RA/M6348/M6220/M8024/8024/8024F/M8024-k switches. Dell PowerConnect series switches closely conform to networking industry standard operational capabilities and administrative interfaces. The differences below should be studied carefully as attempting to configure or operate the PowerConnect switches in the same manner as for previous releases of firmware for PowerConnect 8132/8164/8132F/8164F/7024/7048/7024P/7048P/7024F/7048R/7048R-RA/M6348/M6220/M8024/8024/8024F/M8024-k switches may lead to unexpected results.

Release 5.1.1.7

 No Changed Functionality in this Release

Release 5.1.0.1

 IGMP Snooping

IGMP snooping is enabled by default. Traffic addressed to res er ved multicast IP addr es ses is flooded. Unregistered multicast is flooded to all ports in the VLAN until a multicast router port is identified. Once mrouter port is identified the traffic is forwarded to mrouter port and listener ports only.

 PIMSM

PIMSM Rendezvous Points can be positioned anywhere in the network, not just as the first hop router. Although multiple Rendezvous Points can be configured, only one Rendezvous Point is active at any time.

 Auto-Configuration

Auto-configuration recognizes any of the assigned internal switch’s MAC addresses when present in an autoconfiguration file. The switch re-writes the file to use th e base MAC address of the switch.

 Dot1x Clients

The maximum number of 802.1x clients (i.e. supplicants) that can be authenticated per port is increased to 24. This increase does not include a corresponding increase in the maximum number of 802.1x clients that are supported on an entire switch or stack.

Release 5.0.1.3

 No Changed Functionality in this Release

Release 5.0.0.4

 Stacking Ports C hange

The maximum number of ports that can be configured for stacking per switch is limited to 8. This is done to ensure that the Hardware limits for maximum stack trunk members are not exceeded.

 Stacking Over QSFP+ Ports

Stacking is supported over standard QSFP+ ports at either 1x10G or 4x10G mode. The ports must be configured as stacking ports.

14 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

 Bootcode upgrade

Bootcode program is enhanced to automatically upgrade bootcode on migrating switches from pre-4.x version to 5.x version.

Release 4.2.2.3

 No Changed Functionality in this Release

Release 4.2.1.3

 VoIP Phone Limits

The limitation on the number of VoIP phones has been increased to 576 phones for the

7024/7048/7024P/7048P/7024F

switches.

Release 4.2.0.4

 LAG Limits

Ports can be formed into LAGs in a more flexible manner. The system supports up to 128 total LAGs. Up to 144 ports can be assigned to dynamic LAGs. Up to 72 LAGs can be configured as dynamic. A LAG may contain up to 8 ports. The M8024 supports 12 total LAGs (static or dynamic) with up to 24 ports assigned to dynamic LAGs.

 Stacking Over Ethernet Por ts only on M8024-k/8024/8024F

Stacking is supported over standard Ethernet SFP+ ports. The ports must be configured as stacking ports.

 iSCSI Default Changes

iSCSI optimization is enabled by default. iSCSI optimization will reconfigure ports that are attached to Dell EqualLogic arrays to utilize spanning-tree portfast and unicast storm disable.

PowerConnect

Release 4.1.1.9

 No Changed Functionality in this Release

Release 4.1.0.19

 PHY microcode upgrade process

The PHY microcode upgrade process has been enhanced to upgrade the PHY microcode to the latest version base d on PHY revision. If the user experiences problems links on the combo ports after the upgrade completes, a reboot may be required in order to activate the new PHY firmware.

 Asymmetric flow control

Asymmetric flow control is implemented for the PC8024X, PCM8024, PCM6348, PC70XX, and PCM8024-k switches. The switch does not generate pause frames when congest ed. It will honor pause frames as per industry standards.

Release 4.1.0.6

 Authentication

The enable and line authentication methods will no longer perform authentication if a password for the method is not configured. Previously, these methods would always succeed if no password was configured. To achieve the same functionality, add the “none” method to the list after enable or line method.

System Firmware Version 5.1.1.7 Page 15

NOTE: It is recommended that administrators attach the service/out-of-

management of PowerConnect switches over the operational network.

The default authentication list for telnet and SSH has been changed to enableNetList. The only authentication method contained in enableN etList is enabled. The net effect of these two changes is that a password is required to enter privileged exec mode when using telnet or SSH.

Administrators wishing to maintain the previous PowerConnect behavior can set the default authentication list for telnet and SSH to enableList, which has the enable and none authentication methods (no password required to enter privileged exec mode). The following commands ch ange the telnet authentication method to enableList.

console(config)# line telnet console(config)# enable authentication enableList

console(config)# exit

 New Web Interface

The Web interface has been enhanced with new navigation features for ease of use.

 CLI Syntax Changes

The CLI has changed significantly to be compatible with the PowerConnect switch standard CLI. Configurations for previous releases may not be compatible with th is release and may need to be updated. document for more info.

 Unit/Slot/Port Naming Conventions

In-band interfaces are na med based on stack unit, slot, and port. Units range from 1-12. Slots range from 0-2. Ports range from 1-48. Slots for plug-in modules are numbered 1 and 2. Fixed ports belong to slot 0.

The service port is still addressed using the out-of-band keyword.

 Management VLAN Deprecated

The PowerConnect series switches do not have an in-band management VLAN by default. Administrators can designate a VLAN for support of in-band management operations.

 VLAN interface configuration mode enables routing

When executing the “in terface vlan x” command, routing is automatically enabled on that VLAN.

 Service/Out-of-band Ethernet Port Defaults to DHCP Addressing

By default, the service/out-of-band Ethernet port will attempt to obtain an address via DHCP.

Refer Configuration Migration

band Ethernet port to a physically separate network for out-of-band network management. The service port does not offer routing or switching capabilities nor does it offer enhanced protection from DOS attacks. Configure a VLAN on one or more in-band interfaces for

 LACP Ports Inactive Until Attached

Ports in a LAG configured to use LACP (dynamic LAG) remain inactive (discard received traffic) u ntil they become attached to the LAG. LACP ports that are attached to a LAG will enter the discarding state if they become detached from the LAG for any reason.

Port level configuration for a port that is configured in a dynamic LAG is disregarded. Remove the port from the LAG to restore use of the port level configuration.

Ports in a static LAG begin forwarding on link up. Ports in a static LAG disregard port level configuration. Configure static LAG functions on the static LAG interface.

16 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

NOTE: It is recommended that administrators disable portfast and auto-

portfast on physical interfaces configured in a LAG. Portfast and autoportfast can interfere with an interface entering into LAG mode on a reboot and possibly enable a packet sto rm .

 Spanning Tree Changes

Administrators may assign more than 1024 VLANs to MSTP instances. Only VLANs that are configured on the switch will forward traffic.

The PowerConnect swit ches implements the 802.1Q-2005 standard which builds on 802.1D-2004. 802.1D-2004 incorporates the 802.1t, 802.1w and 802.1s revisions. Port path costs are calculated based on the interface speed as shown below and are dynamically recalculated on interface a ctivation and link sp eed changes.

External Port Path Cost values (Port Path Cost in 17.14 of 802.1D-2004) are applicable in STP, RSTP, and MST modes (Ref. Table 17-3 802.1D-2004). Use the spanning-tree cost command in interface mode to set the external port path cost.

Link Speed

10 Gb/s 2000 1 Gb/s 20000 100 Mb/s 200000 10 Mb/s 2000000 1 Mb/s 20000000

Internal Port Path Cost values are specific to MST mode only (Ref. Table 13-3 802.1Q-2005). Use the spanning-tree mst <instance> cost command in interface mode to set the internal port path cost.

Link Speed

10 Gb/s 2000 1 Gb/s 20000 100 Mb/s 200000 10 Mb/s 2000000 1 Mb/s 20000000

 User Configurable CLI Banners

Administrators may configure banners for the following: MOTD, login, and exec. The banners may consist of multiple lines of text. Each new line will consume an extra two characters (CR/LF) that count against the maximum lengt h banner that can be configured.

 Captive Portal

Captive portal has been extended to support user logout and localization.

 802.1Q

The following changes have been made to the operation of VLANs.

VLAN Membership:

Default Value

System Firmware Version 5.1.1.7 Page 17

 VRRP

By default, trunk ports participate in all VLANs. VLANs created after a trunk port is created are added to all trunk ports. VLANs deleted are removed from all trunk ports. The operator may configure a trunk port to explicitly disallow certain VLANs.

Native VLAN Configuration on T runk Ports:

It is now possible to configure the native VLAN on a port in trunk mode. Trunk mode ports will accept untagged frames but will always transmit tagged frames except for the native VLAN which will always transmit untagged frames. It is also possible to configure a trunk port to drop untagged frames by filtering on the native VLAN, e.g. by

using the switchport trunk allowed vlan remove command.

A trunk port always has a native VLAN (default is VLAN 1), so the default behavior is that untagged packets are treated as if they are tagged in VLAN 1. To drop untagged packets, configure switchport trunk allowed vlan remove <vlan> which has th e side effect of dropping tagged packets in that VLAN as well IIRC.

Switchport Mode Configuration Preserved:

When switching between switchport modes (access , trunk, and general), the switchport configuration applicable to the selected mode is maintained. This means that when switching from one mode to another and back, the port will have the same configuration as it had in the original mode. Only the configuration applicable to the selected mode is active on the port.

The following enhancements have been made to the operation of VRRP to increase usability and robustness of operation in the network:

Preemption Delay:

Per the VRRP RFC 3768, when preemption is enabled, the backup router discards advertisements until the master down-timer fires. When the preemption delay timer is set to a non-zero val ue and the backup switch r eceives a PDU with a lower priority from the master, then backup switch waits for the preemption delay value be fore advertising itself as the master.

Timer Advertis e me nt Lear ni n g:

In VRRP, all participating routers should be configured with coherent advertisement timer interval values. The operator can now enable t imer learning which causes a backup router to learn the master advertisement interval and change its master down interval accordingly.

Ping-able VRRP Interfaces:

RFC 3768 specifies that a r outer may only accept IP packets s ent to the virtual router’s IP address if the router is the address owner (master). In practice, this restriction makes it more difficult to troubleshoot network connectivity problems.

This capability adds support for responding to pings by the VRRP master, but does not allow the VRRP Master to accept other types of packets. A configuration option controls whether the router responds to Echo Requests sent to a VRRP IP address. When enabled, the VRRP master responds to both fragmented and un-fragmented ICMP Echo Request packets. The VRRP master responds to Echo Requests sent to the virtual router’s primary address or any of its secondary addresses. When the VRRP master responds with an Echo Reply, the source IPv4 address is the VRRP address and source MAC address is the virtual router’s MAC address. The VRRP master does not respond to pings sent from the master.

Members of the virtual r outer who are in backup stat e discard ping packets destin ed to VRRP addresses, just as they discard any Ethernet frame sent to a VRRP MAC address.

Fragmentation and Reassembly:

18 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

NOTE: By default, multicast frames are flooded by the switch. Utilize the

multicast frames.

Fragmentation and reassembly of VRRP packets is not supported.

 DHCP Relay

The following enhancements have been made to the operation of DHCP Relay to bring the implementation into conformance with RFC 4649:

DHCPv6 Relay Circuit Id/Remote I d Types

RFC 4649 specifies the IANA assignment of the Relay Circuit Id sub-option and Remote Id option. The implementation has been changed so that the administrator can no longer assign a numerical value to these TLVs as the IANA assigned number is now used. The administrator can still enable or disable the insertion of these TLVs in messages sent to the DHCP server.

Relay Information Option:

The operator has the ability to enable DHCP Relay Information Options both globally and on a physical interface. The interface configuration overrides the glob al configuration for the selected interface.

Relay Information Option Check:

When DHCP Option-82 insertion is enabled for a rel ay agent, the server should echo received Option 82 unaltered back toward the client. The relay agent is required to strip Option 82 information before relaying the BOOTPREPLY to the DHCP client. When enabled, the Relay Information Option Check will cause the BOOTPREPLY packet to be dropped if invalid sub-options are echoed by the DHCP server.

 L2 Address Table

The administrator can disable MAC address table aging.

The administrator can configure static forwarding of a MAC address on a specific VLAN.

mac address-table multicast filtering command to disable flooding of

 LLDP Enhancements

Multiple Neighbor Support:

Multiple neighbors are supported on a single LLDP interface. The number of recognized neighbors is limited to two per port or 834 LLDP neighbors on a fully stacked set of switches. There is no restriction on the number of neighbors connected to an LLDP port. If more LLDP neighbors are pr es ent than are supported, then only the last two neighbors that communicate with the local LLDP interface are recognized and any addit ional neighbors are ignored.

EEE Support:

Support is added to process/communicate the EEE TLV to partner devices. The EEE TLV is an 802.3 organizationally specific TLV used to r eport on the EEE Data Link Layer capabilities.

LLDP-MED Support:

LLDP-MED uses LLDP’s organizationally specific TLV extensions and defines new TLVs which make it easier to deploy VoIP in a wired or wireless LAN/MAN environment. The LLDP implementation supports the following TLVs:

System Firmware Version 5.1.1.7 Page 19

Mandatory 802.1AB TLVs

• Chassis ID TLV (subtype shall default to MAC Address)

• Port ID TLV (subtype shall default to MAC address

• TTL TLV

• MAC/PHY configuration/status TLV

• End of LLDP PDU

Optional 802.1AB TLV

• Systems Capabilities TLV

• Po wer via MDI TLV

NOT recommended

Mandatory LLDP-MED TLVs

• LLDP-MED Capabilities TLV

This TLV allows the network connectivity device to definitively determine wheth er particular connected devices do support LLDP-MED and to discover which specific LLDP-MED TLVs the particular end point devices are capable of su pporting as well as what specific device class they belo ng to.

• Network Policy TLV

This TLV allows the device to advertise its VLAN and associat ed Layer 2 priority and Layer 3 DSCP attributes which apply for a set of specific protocol applications on this port.

• Location Identification TLV

This TLV provides the advertisement of location identifier information Class II endpoint Devices. This is expected to be related to wire map or similar network topology data, such that the configuration of the network Connectivity device is able to uniquely identify the physical location of the connected MED endpoint.

• Extended Power-via-MDI TLV

This TLV allows for advanced power management between endpoints and network connectivity devices. It transmits fine grained power requirement detail s . This TLV provides significantly more value than the

802.1AB Power via MDI TLV.

• EEE TLV

The EEE TLV is used to exchange information about the EEE Data Link Layer capabilities. Devices that require longer wake up times prior to being able to accep t data on their receive path s may use the Data Link Layer capabilities to negotiate for extended system wake up times from the transmitting link partner. This mechanism may allow for more or less aggressive energ y savi ng modes.

 D yna mic V LA N As sig n me nt

Dynamic VLAN assignment is intended to support the connection of hosts to a router with enhanced levels of service, typically either security or QoS. This release supports dynamic VLAN assignment as assign ed from the RADIUS server as part of port authentication. The following additional checks are performed in support of dynamic VLAN assignment:

Before assigning the port to RADIUS assigned VLAN, dot1x checks if the given VLAN is in the VLAN database or not. If the assigned VLAN is not in the VLAN database and dynamic VLAN assignment is enabled , a VLAN is created on the port over which the client is authenticated. Each time a client is de-authenticated on an interface with a particular VLAN, a check verifies if there any other interface which a VLAN member is . If there is no interface as a member, the VLAN is deleted. This behavior is same for MAC b ased authentication as well.

 Usability Enhancements

In the output of the show running-config command, the slot and member configuration is commented with the switch/slot type in human comprehensible form.

When in interface config mode, CLI users can navigate to a different interface b y enter ing the appropriat e interface command without leaving interface config mode.

CLI users can log out of the switch using the exit command (exit is an alias for quit). The CLI Reference Guide is updated with acceptable character sets and maximum length s fo r s tring parameters to

commands. Management ACLs permit specification of service any as shorthand for enabling all services access for in-band

management.

for transmission in order to conserve LLDPDU space.

20 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

VLANs may be administratively assigned to MSTIs in excess of the switch physical limits and without regard to whether the VLAN is actually configured. Frames are only forwarded on VLANs assigned to interfaces.

Administrators can re-enter SYSLOG server config mode for a particular SYSLOG server entry without requiring the deletion and re-creatio n of the entry.

Administrators can configure the web timeout by navigating to: System -> Management Security -> Telnet Server -> Telnet Session Timeou t.

User configured banners (login, exec, MOTD) appear in the running config. By default, auto-install supports image downgrade for network installs, specific version USB installs (using a .setup

file), and stack firmware s ynchronization. A comprehensible message and recommendation is issued when configuring multiple services (telnet, http, etc.) to

listen on the same TCP port. The terminal length command allows user control over terminal paging.

 Simple Mode

The PowerConnect M8024-k is the only modular switch that defaults to the simple mode of operation. Simple mode contains a restrict ed set of commands suitable for control of a port aggregation device that can be deployed in a network without requiring updates to the network by a network administrator. Users needing switch capabilities which require the network administrator to modify the network configuration can exit simple mode using the no mode simple command.

 AAA Authentication

In prior releases, more than one method could be specified for dot1x authentication even though only the first method was attempted. The CLI and Web now only accept a single method for dot1x authentication.

System Firmware Version 5.1.1.7 Page 21

Summary

User Impact

Resolution

Affected Platforms

All 5.1 supported

Stack member crash with

Switch crash occurs

Corrected the SDM Templat e

PC8132

Stack member crash when

Switch crash occurs

Corrected a stack data plane

PC8132

PC8164F

Stack member crash from

Switch crash occurs

Enhanced DMA retry logic to

PC8132

Stack member crash in

Switch crash occurs

Corrected multicast routing

PC8132

PC8164F

PC8132

PCM6220

PCM6348

All 5.1 supported

PC8132

PC8164F

PC8132

PC8164F

All 5.1 supported

Issues Resolved

The following issues from previous releases have been corrected. The issues listed here may have been discovered on any of the switches listed on the title page.

Release 5.1.1.7

Switch crashes in tIomEvtMon() task once per week.

Switch crashes in tCptvPrtl task.

“SDM Template mismatch” error

“ISO count mismatch has been detected”

DMA hang

mcastMapTask

Stack master can move on stack reload

iomEventLog() might result in switch crash

FRU service tag value up dates from CMC might result in buffer overflow as service tag and other related tag fields are updated in the switch.

Switch crash occurs in captive portal after a reload whi ch brings down the entire switch

occasionally when an SDM Template id mismatch occu rs.

occasionally when th e stack unit detects that it is isolated.

occasionally when packet DMA from the switching fabric to the CPU stops.

occasionally with routing and multicast traffic.

On a stack reload the st ack master can move to a different switch based on the plug-in modules being used.

Communications error in the m1000e Chassis CMC can result in a crash.

Correct buffer overflow.

Corrected captive portal initialization problem

ID problem so that invalid SDM template ids will be ignored.

communications problem.

allow DMA status information to update completely.

problem.

Corrected stack master timing problem when 10GBaseT plug-in modules are used.

Corrected index r a nge error on data received from the CMC

PCM6220 PCM6348 PCM8024

PCM8024-k

platforms

PC8132F PC8164 PC8164F

PC8132F PC8164

PC8132F PC8164 PC8164F

PC8132F PC8164

PC8132F PC8164 PC8164F

PCM8024 PCM8024K

Switch crash in DHCP server task

Stack member crash in bcmRLINK task

Stack member crash in DMA processing

Port detection mechanism fails with SNMP Query using Q-BRIDGE-MIB

22 System Firmware Version 5.1.1.7

Switch crash in DHCP server task

Switch crash in bcmRLINK task due to unit id being out of range

Switch crash in DMA processing due to memory corruption

dot1qVlanFd bId S NMP object returns incremental indexes of the VLANs.

Corrected memory corruption problem.

Corrected memory corruption in ATP process.

Corrected in problem in cache coherency which was causi ng data corruption.

Corrected problem so that the object will return VLAN IDs

platforms

PC8132F PC8164

platforms

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

All 5.1 supported

MIB file parsing errors

Errors related to undefined

Corrected parsing errors.

All 5.1 supported WebUI config issues with

Copy-from/Copy-to WEB UI

Corrected copying for duplex

All 5.1 supported

Trunk port tags traffic after

All 5.1 supported

All 5.1 supported All 5.1 supported

PC70xx returns the chassis

Corrected default value for

All PC70xx platforms

PCM6220

PC8024

Need to throw

All 5.1 supported All 5.1 supported

PC8132

Incorrect stats for "show logging email statistics"

Telnet to port 80 hangs management

Switch Hangs after using Web GUI -> SNMP -> Access Control

"No of email Failures so far" in command "show logging email statistics" is incrementing even if there are no failures.

Console hangs when telnet session to localhost is initiated

Switch Hangs when navigating to SNMP -> Access Control -> Show All page

Corrected problem in checking for non-urgent messages.

Corrected task communication problem in telnet tasks.

Corrected problem accessing access control data structure

platforms

pre-provisioned por t s in stacked environment

upgrading from v3.x firmware

Web UI issue attributed to ip local-proxy-arp setting

Stack fail-over causing loss of access to ESX VM environment

service tag as N/A instead of NULL

IPv6 VLAN interface fails to handle size specification

objects in 802.1ag MIB ar e displayed when snmpwalk is performed on nets nmp software.

functionality doesn't work properly on IE 8

Switchport mode trunk & switchport trunk commands changed from previous firmware version

Attribute to ip local-proxy-arp enabled by default on routed VLANs if WebUI is used to configure.

Switch will send tagged traffic out of the ports in native vlan.

service tag as N/A and it causes issue with network management system applications.

ipv6 ping fails when the size parameter is specified

parameters

Corrected configuration migration problem

Corrected default attribute.

Corrected problem in native VLAN check.

chassis service tag.

Corrected packet fragmentation problem

platforms

PCM8024 PCM8024K PCM6348 PC70xx

ERROR/Warnings on switch when PIM SM neighbor exceeds 256 neighbor count

Traffic loss after restarting traffic after power cable pull/insert, high CPU utilization.

OOB interface loses the static default gateway configuration after multiple failover

System Firmware Version 5.1.1.7 Page 23

PIM SM neighbor count gets beyond 256 random neighbors gets deleted without any error or warning messages.

After restarting multicast traffic for hard power cycle of a switch in the network there may be multicast data lost.

Out of Band interface state is down during initiate failover event causing default gateway configuration failure.

Add validation such that no new neighbors are entertained if the router maximum neighbor capacity is reached.

Corrected problem in event notification logic.

Corrected link state check in the service port.

platforms

PC8132F PC8164 PC8164F

M8024-K switch can cause

Temp range difference between

Modified fan speed change

PCM8024-K All 5.1 supported

PC8132 Routing issue, null address

ARP replies destined to the

Corrected problem when

All 5.1 supported

With Traffic Running,

The ping packets destined to

Corrected ping priority

All 5.1 supported

Summary

User Impact

Resolution

Affected Platforms

M6220 crashes when

Switch crashes when using this

Fixed memory issue in a loop

All 5.0 supported

All 5.0 supported All 5.0 supported

Incorrect SFP interface log

The wrong port may get

Calculate the correct i nt ernal

Platforms that support

transceiver modules

All 5.0 supported Simple mode In-band IP

"ip address vlan" command

Put "ip address vlan" after

PCM8024-k

ARP entries are purged for

L3 egress objects are not

Use the SDK init function to

All 5.0 supported

high fan speed with CMC

4.2

All entries in IGMP snooping are added and deleted immediately in a specific time for 5 minutes

Manager is rebooted while doing clear config due to crash in mcastMapTask.

in ARP table after VLAN manipulation

cannot PING VRRP IP from VRRP backup router

Release 5.1.0.1

M8024k and CMC with FW 4.2 and above causing chassis fan speed stuck at 100%

Group entries in IGMP snooping are added and deleted immediately in a specific time for 5 minutes

During multicast traffic flow, on initiating a NSF failover, the unit acting as standby crashes and there will be traffic lo s s.

DUT L3 MAC address will be dropped.

VRRP Master IP are coming in the same CPU CoS queu e as the unresolved data packets. Since these ping packets are not prioritized, they are getting dropped at the CoS queues itself resulting in ping to fail

parameters and added new temperature handling logic.

Corrected timer rollover problem that happens every

49.71 days

Corrected data copy pro blem when check pointing MFC data.

removing ARP table entries

problem.

platforms

PC8132F PC8164 PC8164F

platforms

issuing command “show interfaces switchport po1”

Cannot apply ACL on VLAN 'out bound' direction from GUI

Default VLAN cannot be made static from GUI

messages on stack

SNMP management IP address can only be set from WebUI and lost after reboot.

address missing after reboot.

unknown reason

command

User needed to use CLI Fixed the issue to be able to

apply in both directions from the web

Cannot change VLAN member ports settings

reported in the trap notification

command is missing in the CLI in Simple Mode

disappears from the runningconfig after reboot

programmed correctly resulting in incorrect L3 forwarding.

Fixed VLAN membership web page

interface number

Added CLI command support

"port-aggregator group" commands in the text config

set defaults for egress object.

platforms

XFP, SFP a n d SFP+

platforms

PCM6220 PCM6348 PCM8024

platforms

24 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

All platforms that support

All 5.0 supported

Get the mask from the

All 5.0 supported

M8024-k, M8024, M6348

CLI commands are not

Enabling password recovery will

Fixed the issue to enable

All 5.0 supported PC8024 and PC81xx

Check if any weight is

Platforms running DCBX Switch GUI forcing 100Mb

Switch GUI set all external ports

Corrected port speed

M6348 PowerConnect

Vlan membership information

Javascript fixed to pick correct

M6348 email addresses with

A valid email id with an

Underscore is now considered

All 5.0 supported

OOB Static IP unreachable after stack failover

http[s] authentication against RADIUS only allows privilege level 1 access

SNMP v1/2 community manager address no lon ger functions with network address.

Switch service tag not displayed by “show system id” command

After failover stack loses static IP address configured on OOB interface.

HTTPS authentica t ion through RADIUS grants only read-only access.

When a subnet IP is set as SNMP community IP addr es s then snmp requests from hosts of that subnet are not accepted on the box.

Switch service tag not displayed with “show system id” command

Proper checks prevent using the previous DHCP mode.

parse RADIUS server respo nse properly

community IP address configured and set it instead of the hard coded value.

Added retry to get the Service Tag value

the OOB interface

platforms

authorized after Passwor d Recovery

LINK UP on all interfaces during POST,

FCoE -M8024-k setting TSA map to link strict and assigning bandwidth allocation to TC

speed on External ports when cloning port configurations

M6348/General port GUI issue

M6348 - service tag not displayed in GUI

underscore "_" are reject ed phone port configuration

macro incomplete

not allow the user to run CLI commands on the serial console.

Links are UP and flapping during power reset and reloads.

In the CEE mode there is no way to specify the TSA mode in the ETS TLV

to 100Mb speed when cloning port configurations.

on web is displayed and applied incorrectly

Service tag shows up as none on GUI

underscore cannot be used The macro never returns and

appears to hang

commands from serial console also

Changed the CPLD and reset logic

configured for the TCGs, if so set the mode to ETS.

processing from WEB

enum values

Synchronization issue fixed to retry and get the tag information

valid character Fixed the issue and return error

platforms

and including ETS

M6348

platforms

System Firmware Version 5.1.1.7 Page 25

Release 5.0.1.3

Summary

User Impact

Resolution

Affected Platforms

Re-enable Auto

The Auto Negotiation can't be enabled back

Corrected error in Port

All 5.0 supported VLAN ACL

members

ACL prevented traffic from reaching

Prevented VLAN ACL from

All 5.0 supported All 5.0 supported

All 5.0 supported

All PC70XX Switches

All 5.0 supported

Change PoE

PoE controller Disconnect Type was

PoE controller Disconnect

PC7024P All 5.0 supported

All 5.0 supported PC8132

PC8164F

Negotiation using GUI doesn't work properly

blocks traffic across stack

Power supply logging

'no ipv4 or ipv6 address found in response for request id' error

phone port configuration macro incomplete

web session timeout

RX Equalizer setting on Stacking ports.

interface speed options

on "Port Configuration->Show All" page once it has been disabl e d.

destination

Power Supply messages are only sent to the log file.

'no ipv4 or ipv6 address found in response for request id' error message flooding the switch

The phone port is missing some needed commands

Web session would timeout regardless of the timeout setting

RX Equalizer was not set on the stacking port which caused stacking errors to be reported.

Invalid speed option s mess age i s not clear Changed the message repo rted

Configuration page.

being applied to stack ports.

Elevated failure messages to ERROR severity

Corrected problem that was causing extra error messages to be displayed.

Added missing commands to the pone port macro.

Corrected the hard t imeout to be one hour.

Added RX Equalizer value.

when an invalid speed option is input.

platforms

Disconnect Type to DC disconnect

stacking with extension modules

DHCP Req pkts are not being forwarded to Voice VLAN Component

Netlogic memory and B0 CPU support

changed to DC Disconnect.

Stacking with redundant stack rings caused packet flooding on the stacking links.

When using Avaya phone on a switch running dot1x, the phone will NOT get authenticated and thus will not boot

Netlogic memory and B0 CPU support Added support for Netlogic

26 Syst e m Firmware Version 5.1.1.7

Type was changed to DC Disconnect.

Corrected problem with st ack link queuing.

Corrected DHCP forwarding problem.

memory and B0 CPU.

PC7048P

platforms

PC8132F PC8164

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Resolution

Affected Platforms

OpenManage UI

Port description does not accept space

Correction the por t des c r ipt ion

All 4.2 supported

"show spanning-

The user may think that th er e is traffic over

Corrected the counters.

All 4.2 supported

using the web.

All 4.2 platforms that

All 4.2 supported "encapsulation"

The user could only set VLAN

Added support for the

All 4.2 supported M8024-k

The user could not get accurate information

Corrected the information

PCM8024-k

"spanning-tree

The user could not see the configuration of

Corrected the output of the

All 4.2 supported Inconsistent

The user may be able to configure an

Corrected the error handling so

All 4.2 supported

Release 5.0.0.4

does not accept blank space character for port descripti on

tree detail" counters are non-zero for inactive interfaces

Unable to remove 0.0.0.0 as default gateway on OOB interface

When connecting via SSH, the user is not prompted to acknowledge the MOTD

command is unavailable in interface range mode

interfaces that are supposed to be inactive as far as spanning tree is co ncerned.

The user may not be able to remove the default gateway for th e O O B interface causing unwanted traffic to go over the OOB interface.

The user may log in without having acknowledged the MOTD.

encapsulation on a sin gl e VLAN at a time.

validation to accept spaces.

Added support for setting the gateway to 0.0.0.0.

Add functionality to send the acknowledge question when new SSH connection is created, but after authentication.

encapsulation command in interface range VLAN mode.

platforms

support the OOB interface

platforms

OpenManage Web UI stack view display

transmit holdcount" command's value is not reflected in show command

behavior in single and range interface mode of "spanningtree port

CLI command "no passwords strength excludekeyword" requires <keyword>

System Firmware Version 5.1.1.7 Page 27

about a stack of M8024-k switches from the stack view web page.

the hold-count parameter.

invalid spanning tree port priority in range mode.

The user could not reset the excluded keywords to the factory default of no excluded keywords with a single command. It was necessary to remove each excluded keyword one at a time.

displayed on the stack view web page.

"show spanning-tree detail" command to include the configuration of the hold-count parameter.

that the invalid priority is not used.

Correctly implemented the "no" form of the "passwords strength exclude-keyword" command.

platforms

Inline help for

The help text might lead the user to think

Corrected the help text so that

All 4.2 supported

Some of the

Some of the options of "show ip pim"

Added missing options to the

All 4.2 supported

Configuration

Configuration command "logging

Added checks to return error

All 4.2 supported

dot1dTpPortMa

dot1dTpPortMaxInfo is displaying

Excluded the MAC header

All 4.2 supported

If a user

No error message was given.

Added a message once an exit

All 4.2 supported

"spanning-tree loopguard" is incorrect.

options of "show ip pim" command are not available in user EXEC mode.

command "logging <hostname>" accepts more than 63 characters.

xInfo is displaying maximum frame size that includes MAC header.

attempts to create a certificate request with information that is not identical to the key generated, the user is not gi ve n an error/ informative response letting them know there is a conflict between their key and their request.

OIDagentInvent orySupportedUn itExpectedCode Ver provides incorrect display string 1.0.176.0

the command only applied to a single port instead of all ports.

command are not availabl e in user EXEC mode.

<hostname>" accepts more t han 63 characters, but saves only 63 characters in running-config.

maximum frame size that includes MAC header.

agentInventorySupportedUnitExpectedCode Ver object returns incorrect values.

it indicates that the command applies to all ports.

CLI tree.

when hostname is larger than 63 characters.

length when retrieving the OID information

is issued from the request or generate command. This is to tell the user whether th e operation was successful or not.

Made the OID that is in p rivate MIB obsolete as it's not required.

platforms

28 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

All 4.2 supported

CLI command

The user would not always get the correct

Corrected the logic in the

All 4.2 supported All 4.2 supported All 4.2 supported

All 4.2 supported

1. Changes are not applied to the correct interfaces on the pages "System>sFlow>Sampler Configuration" "System>sFlow->Poll Configuration"

2. Incorrect error message is displayed when not configured Receiver index is applied to interface on "System>sFlow->Poll Configuration" page.

"show ip route configured ...longerprefixes" reports that default gateway is not configured, which is incorrect.

1. Changes were not applied to the correct interfaces on the GUI pages referred to.

2. An incorrect error message wa s displayed.

The default gateway would not be shown on executing "show ip route configured"

1. Used initialized buffers.

2. Correct the error messag e.

Corrected the implementat ion of this command so that it functions correctly.

platforms

"show ip route

192.168.2.2 /24" not ignoring host bits.

CLI command "show spanningtree detail" missing information.

"ipv6 pim joinprune interval" command does not work

Macro input string does not accept more than 20 characters

System Firmware Version 5.1.1.7 Page 29

list of routes from the "show ip route" command.

The user could not see the configured values for spanning tree max hops.

The user could not change the configuration of the IPv6 PIM join interval.

The interface string length is lim ited to 20 characters for the "macro global apply" command.

implementation of the "show ip routes" command so that the appropriate list of routes are shown, depending upon which optional parameters are given.

Changed the output of the "show spanning-tree" and "show spanning-tree detail" commands to include the configured value of max hops.

Corrected the implementat ion of this command so that it functions correctly.

Increased the acceptab le string length for the "macro global apply" command to 256 characters.

platforms

startup-config

startup-config failed error message is

Corrected the implementat ion

All 4.2 supported In stacking

CLI output for command "show system

Corrected pagination in the

All 4.2 supported PC8024F

Alignment

The alignment issues make it d ifficult for

Corrected alignment issue

All 4.2 supported All 4.2 supported All 4.2 supported

All 4.2 supported

"show interface

The user sees the info but it looks messy

Corrected formatting i s s ues

PC8024

All 4.2 supported

Port channel

The members of the stack other than the

Corrected the logic to check

All 4.2 supported

failed error message is displayed when finishing the Dell Easy Setup Wizard

environment, "show system temperature" command output is corrupted

Error message for port enable / disable on IE on 8024F combo ports

issues in the output of "show voice vlan interface" command

CLI command "show ip igmp snooping" output is truncated.

vrrp ip command accepts invalid IP addresses

CLI command "isdp enable" does not work for a range of ports

displayed when finishing the Dell Easy Setup Wizard

temperature" is corru pted when pagination is used.

When a copper port was disabled, port was getting set in no negotiating mode.

the customer to read the page.

The displayed info is not complete on page. Corrected the implementation

Switch allowed user to misconfigure VRRP IP

User cannot enable isdp on range of interfaces

of this command so that it functions correctly.

command handler.

Corrected by disabling negotiation on combo ports

of this command so that it functions correctly.

Added checks to return error when invalid VRRP IP address is given.

Added support to "isdp enable" command in interface range mode

platforms

Priority-FlowControl" command output is not formatted correctly

CLI command "show ip route

192.168.2.2 /24" not ignoring host bits.

goes down during master failover.

30 System Firmware Version 5.1.1.7

and is hard to read.

The user would not always get the correct list of routes from the "show ip route" command.

master unit and thei r por t channels, i.e. the Standby units and their port channel interfaces, are now not affected by a master failover and the port channel stays up during failover.

Corrected the logic in the implementation of the "show ip routes" command so that the appropriate list of routes are shown, depending upon which optional parameters are given.

for active port channels in the new Standby master unit.

PC8024F PCM8024-K

platforms

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

switches

emWeb

The switch crashes when it receives the url

Corrected URL length problem

All

general mode

vlan

The general mode for switchport won't

Corrected memory

All SNMP OIDs

switch crash

Walk of any member of

Corrected unit number range

All Stack ARP

ARP replies to a routing interface are not

interfaces.

Corrected ARP repl y policy

All

Stack member

GUI.

Stack member unit ports were not displayed

Corrected power LED element

PCM8024-K

No response to

jumbo (ping)

Third IP fragment was being dropped in

Corrected jumbo Ping packet

All

Ports connected

after reboot.

When customer is using the Intel X520

Corrected port configuration

PC8024

switch reboots randomly

Corrected memory corruption problem.

All

Summary

User Impact

Resolution

Affected Platforms

IP routing

enabled.

Traffic will get forwarded by software when

Telnet existing

"New Telnet Sessions" are changed to

Disabled.

OOB interface

Out-Of-Band interface will be unreachable

PCM8024-k

rebooting possible crash dump w/

vlan does not show up after a reload, it shows up in access

1.3.6.1.4.1.674.1

0895.3000.1.2.1

10.7.x causing

Problem

units port display is missing in web

ICMP with

with the maximum characters.

work properly in several cases , such as, script apply, nsf failover and save and reload.

envMonFanStatusEntry and envMonSupplyStatusEntry with 67109251 index causes crash.

received at the CPU after cl ear ing the config and reconfiguring the routing

properly.

jumbo Ping packe t.

initialization problem

check.

problem

id name problem which was causing this display problem.

problem.

to Intel X520 NIC do not return online

NICs, sometimes the ports never came back up after a reboot of the switch.

Release 4.2.2.3

unexpectedly after reboot despite its not having been

session is interrupted when "New Telnet Sessions" is set to "Block".

unreachable after failover

global routing is not enabled.

Block, All the current telnet sessions are closed and using "Telnet Ser ver Admin Mode" to set Disable will only block new telnet sessions.

after a stack failover

timing issue.

Correct the default act ion so that global forwarding is enabled for the hard ware.

Functionality was corrected so that "New Telnet Sessions" will only block new telnet sessions and "Telnet Server Admin Mode" will close all the sessions if the mode is

Corrected MAC address for new manager.

PC8024F

All 4.2 supported platforms

PCM6220 PCM6348 PCM8024

System Firmware Version 5.1.1.7 Page 31

Switch CLI help

Warning message printed during execution

mode only.

Timeout during

root

SNMP walk on root ends with timeout

"show interfaces

value

"show interfaces switchport" command

Error while

access-group

Error message when trying to apply ACLs

Walk of

VLANs

Walk of agentDaiVlanStatsTable returns

OOB interface

When the Setup Wizard is used the OOB

Unable to create

Web GUI

Unable to create new VLAN with its name

"Failed to get

messages in logs

"Failed to get CPU cosq 0 drop counters,

Default "ip

error repeatedly

Default "ip address dhcp" command on agentPortSpeed

error

Mistakes in the

ifIndiscards is

is in trunk mode

ifIndiscards is still counting VLAN discards

does not locate the 'initiate failover' command properly

SNMP walk on

switchport" command shows incorrect General Mode Tagged VLANs

adding an access-list to an

agentDaiVlanSt atsTable returns statistics for all the 4096

configured with default IP via Setup Wizard though not to do

of the deprecated "movemanagement" command incorrectly states that "initiate failover" command is executed from privileged EXEC mode whereas it's available from stack config

shows incorrect General Mode Tagged VLANs value.

to VLANs.

statistics for all the 4096 VLANs even they do not exist in the system

interface IP address is incorrectly set to

192.168.2.1 static IP by default

Corrected the warning message All 4.2 supported

platforms

Corrected delay that cau sed timeout.

Corrected the display of the General Mode Tagged VLANs value.

Corrected the condition that resulted in the error.

Corrected the VLAN Stats Table data.

Corrected the settin g of the OOB interface IP addr es s .

All 4.2 supported platforms

PC8024 PCM8024 PCM8024-k

All 4.2 supported platforms

VLAN ID and name VLAN in

CPU cosq 0 drop counters, error -16"

address dhcp" command on VLAN 1 causing "failed to bind socket"

DuplexStatus SNMP MIB object documentation

still counting VLAN discards if the interface switchport mode

using the Web GUI

error -16" messages in logs

VLAN 1 causing "failed to bind socket" error repeatedly

agentPortSpee dDuplexStatus SN MP MIB object description

if the interface switchport mode is in trunk mode (acceptable frame type is all)

Corrected the VLAN creation GUI

Corrected the retri eval of cosq 0 counters.

Corrected DHCP socket problem.

Corrected MIB object description.

Corrected ifIndiscards counter collection.

All 4.2 supported platforms

32 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Switch doesn't

TLVs.

Switch doesn't register all the 576 devices if

SNMP and LOG

ifAdminStatus

When setting ifAdminStatus “adminState

Cannot type

reboot of switch

After reboot external ser ial connection can

Web GUI

Explorer 9

Device View doesn't display one port on IE

Switchport

running config

"switchport general allowed vlan add 1 Cannot rename

Cannot rename vlan1 running v4.2.1.3.

SNMP Port

When setting VLAN un-tagging for a DNS client error

rebooted.

Upgrade of 3.1.4.5 to 4.1.0.6 DNS client Terminal Length

4.1

Terminal length is not setting per-session is

Trunk port

reboot

Trunk port multiple VLAN assignment FIP snooping

script apply.

FIP snooping session i s not getting

LLDP-MED has all the optional LLDPMED Location TLVs.

Corrected the allocation of LLDP-MED Location TLVs.

All 4.2 supported platforms

errors when setting

input into external serial connector after

doesn't display port 24 if using Internet

general configuration, VLAN1 tagged not saved in

vlan1

tag/untag issue

2d is not valid” errors are produced.

become inoperative.

9. (IE9 is currently unsupported)

tagged" is not displayed in running config

It is correct that the defaul t VLAN cannot be renamed (same as a Cisco switch).

specific port using the dot1qVlanStaticUntaggedPorts object, All the other ports are automatically added to that particular V LAN as tag ged.

Corrected error check when using invalid testing options.

Corrected initialization of serial connection.

Corrected problem for IE9. All 4.2 supported

Corrected check of enabled VLAN.

Updated error message to reflect this behavior.

Corrected the port un-tagging set.

All 4.2 supported platforms

PCM6220

platforms

All 4.2 supported platforms

in logs and switch locks up and needs to be

setting not working as in

multiple vlan assignment doesn't work properly after

session is not getting established after

System Firmware Version 5.1.1.7 Page 33

error “osapiSocketRecvFr om returned error “ in logs and switch locks up and needs to be rebooted.

an enhancement scheduled for the next release.

doesn't work properly after reboot.

established after script apply.

Corrected service por t link status during upgrade.

Corrected a problem with terminal scrolling.

Corrected command pars ing problem.

Corrected FIP snooping command problem.

All 4.2 supported platforms

"ip http secure-

build.

ip http secure-server command not getting Data loop in

A data loop occurs in the port-aggregator if

Summary

User Impact

Resolution

Affected Platforms

The “show fiber

up.

The “show fiber optical-transceiver" Unable to ping

MTU size configured on the combo ports

cpCaptivePortal

The description of the

Corrected the description in

LLDP

TLV

When Port-Description is set as no

The CLI shows

The "show interface advanced firmware" Can't assign a

Config migration for the "name <vlan

command.

No default route

route table

"Show ip route" commands will not display

terminal length 0.

CLI command

CLI command "show dot1x users" is

command.

server" command not getting migrated from 4.1.0.19 build to 4.2.1.3

Simple mode when adding VLAN

applied when we migrate from 4.1.0.19 to

4.2.1.3 release

a new VLAN is added to an interface.

Release 4.2.1.3

opticaltransceiver" command shows LOS even though link is

with jumbo frames set

command was reporting incorrect values when optical transceivers that did not support diagnostics were used.

will be lost after a save and reload and also when changes the media p r eferences are made.

Corrected command pars ing problem.

Corrected the VLAN creation error.

Correct diagnostics so that diagnostics for unsupported transceivers are not d isplayed.

Corrected the storing of the MTU size

All 4.2 supported platforms

PCM6220 PCM6348 PCM8024 PCM8024-k

All 4.2 supported platforms

PC72xx and PC80xx platforms

WebLangCode.

1.1 display s en when mib says only supported value active (1)

Assignment of port ID for PortDescription

incorrect media type 10GBASET for fiber ports

name to a VLAN

or static showing in ip

"show dot1x users" is missing

cpCaptivePortalConfigWebLangCode object doesn't correspond to the values it returns.

description in LLDP port configuration, the TLV should contain the Port Interface name as the port description by default instead of

shows incorrect information under "Type" column.

name>" command was not correct.

route even though there were routes is in the router(static routes, ospf routes) for

missing

the fastpath_captive_portal.mx file

Corrected the default P ortDescription.

Removed the Type column which was not valid.

Corrected the config migration for the "name <vlan name>"

Corrected the display of the "Show ip route" command.

Corrected the command tree for the "show dot1x users"

All 4.2 supported platforms

34 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Missing port

EgressPorts)

Changing VLAN configuration for general

The VRP P track

The VRRP track port priority in the running

Firmware won't

logging email

The hyphen is part of the allowed characters

Internal ports

Internal ports were up while switch is

Corrected the initialization of

Continuous log

default settings

POE log message comes up during power

DNS client error

The meaning of the DNS error message:

unclear.

VRRP

Issues

Ping to a remote host will not work if

VLAN

GUI

VLAN membership port names are not

Radius crash

Sometimes switch crashes when receiving platforms

In a Stack, the

breaks CLI

CLI output for command "show system QOS on port

There is no way to see the match packet

channel number > " was added.

membership from SNMP (dot1qVlanStati cUntaggedPorts and dot1qVlanStatic

mode affected trunk mode configuration and there from the port membership returned via SNMP.

Corrected the issue. All 4.2 supported

platforms

port priority changes in the running config

allow domain names with hyphen "-" in

are up during most of the switch POST

message at

config is retrieving the operational value instead of the configured value.

available for the email address.

booting, this was causing traffic loss in network.

up sometimes.

“DNS Client: osapiS ocketRecvFrom returned error for addr 0x1214BCA8” is

Corrected the retri eval of the configured value

Added the hyphen to the allowed characters for email addresses.

the internal ports. CPLD Code Update is required.

1. Update the CPLD using the command 'dev cpldUpdate' Console#dev cpldUpdate

2. Power cycle the switch. (Power Cycle from CMC WebUI. Do not run “reload” from the console prompt)

Removed unwanted message. PC7024P/PC7048P

Corrected the text o f the message.

All 4.2 supported platforms

PCM8024-K

All 4.2 supported platforms

Intermittent Connectivity

membership port names not consistent in

"show system temperature" CLI command

channel

System Firmware Version 5.1.1.7 Page 35

switch acts as VRRP master

consistent.

of Radius packets

temperature" is corru pted when pagination is used because the stack displays much more data.

counts of a policy-map on a port-channel.

Corrected a VRRP communication issue.

Corrected the port numbers displayed in the GUI.

Corrected the crash. All 4.2 supported

Corrected the paginat ion issue. All 4.2 supported

Command "show policy-map interface port-channel <port-

All 4.2 supported platforms

platforms

All 4.2 supported platforms

Dropped VLAN

Customer is really not able to use this

PC7048 Combo

Combo ports are randomly not transmitting

Order of

changed

The "show running-config" command

Summary

User Impact

Resolution

Affected Platforms

PowerConnect

specification

sFlow implementation is not per

Using config

console

Doesn’t affect any iscsi traffic. Constant log

PC8024

8024 stack does

If a user logs in via console - motd and

All 4.2 supported

Simple Switch

default

FC not functional

All 4.2 supported

OSPF error

With 128 IPv4 & IPv6 interfaces & OSFP

routing.

All 4.2 supported

frames are included in Discards counters.

ports are not passing traffic.

switchport commands in running-config

counter to monitor their network outage activity.

traffic, but receiving frames when media type preference set as SFP

shows the detailed switchport configuration before the switchport mode. This caused problems with some scripts.

Release 4.2.0.4

sFlow implementation is not per

wizard to setup 8024 causes iscsi error messages off out-of-band in

specification

messages with the erro r “Can’t create static rule”.

Corrected by not ignoring the dropped VLAN frames in the Discards counter

Corrected by adding a property to bypass the lane initialization when media-type preferences is set.

Corrected the orde r of runningconfigurati on output.

Issues corrected and passes conformance test

Issue is corrected.

All 4.2 supported platforms

PC70xx

All 4.2 supported platforms

not follow similar banner execution for console, telnet and ssh

should have FC enabled by

message after reboot

login banners are execut ed If a user logs in via ssh - only exec banner is executed If a user logs in via telnet - all 3 banners are executed

configured a reboot caused the following error messages to be gen er ated: Failed to send message to O SPF mapping thread in ospfMapRoutingEventChangeCallBack()

These lost events could lead to incorrect

Behavior is now as follows:

Add motd banner display for telnet connection before motd acknowledge question is asked.

Remove motd banner after motd acknowledge question.

FC is enabled by default in simple switch mode.

Issue has been correct ed.

platforms

36 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

show crypto key

After generating DSA & RSA keys,

There is no impact on the keys themselves.

All 4.2 supported

Web GUI AR P

The Web GUI does not display the

All 4.2 supported

The copper auto

fails to link up

The copper ports 1-24 fail on 8024 to

PC8024

PCM6220 -

page on the GUI

The current stacking port status does not

All 4.2 supported

PC8024 -

interface

If user changes VLAN me mbers h i p

All 4.2 supported

Implement

M8024-K

The dot1qVlanStaticTable is not available

All 4.2 supported

Trunk mode

with Cisco

tagged native vlan traffic.

All 4.2 supported

PCM6220 -

sftp

Switch hangs and needs a hard reboot.

All 4.2 supported

mypubkey generates two error messages

executing the command "show crypto key mypubkey" generates the following error messages:

<187> APR 23 09:09: 22 19 2.168.1.2-1 OSAPI[162603 39 2]: osapi_file. c ( 657 ) 426 %% File close failed for descriptor 37

<187> APR 23 09:09: 22 19 2.168.1.2-1 OSAPI[162603 39 2]: osapi_file. c ( 657 ) 427 %% File close failed for descriptor 37

Resolved

platforms

table does not match the CLI ARP table

negotiation process fails at 100mb FD and

Wrong information for current link status in port configuration

VLAN membership change on interface affects unrelated

dot1qVlanStatic Table in

static/local ARP entries. However, once an external port is linked up some, but not all, of the Static/local appear in the Web GUI.

negotiate at 100Mb FD.

display “stacking linkup”

configurati on on tr u nk port via the WEB he loses ports configuration on access ports and access ports are put into default VLAN.

on M8024-k.

All ARP entries are now shown.

8024 copper ports negotiate to 100M FD.

The web stack display has been updated to accur ately reflect the correct information as well as address some style issues.

Issue is resolved.

the dot1qStaticVlanTable is implemented for the M8024-k

platforms

native vlan implementation not compatible

Switch hangs when user tries to transfer a file from the flash to the server using

System Firmware Version 5.1.1.7 Page 37

Native vlan implementation is incompatible with Cisco. Switch will egress native vlan traffic as tagged and Cisco will drop ingress

Implementation of native vlan changed to be compatible with Cisco.

None – issue is resolved.

platforms

No message

The switch does not log an error message

refused, but no error message is logged.

All 4.2 supported

CLI command

correctly

The CLI command "show ip v6 interface"

All 4.2 supported

2nd and 4th

problems

Ports report to be fiber ports and not copper

PC8024/PC8024F

WebGui Secure

On the WebGui under Management

on the Web UI

All 4.2 supported

Web UI MSTP

MSTIs

In the "MSTP Interface Settings" page of

All 4.2 supported

Flowcontrol

module

Flowcontrol negotiation iss ue s be t ween

PC8024/PC8024F

PC8024 - GUI -

When you change the port setting, for

problem when using the command prompt.

All 4.2 supported

Unable to config

pops up

Unable to configure Secure Shell from web

All 4.2 supported Web interface

iSCSI item.

The web page for iSCSI fails to open when

All 4.2 supported

logged for "Max number of SSH login sessions exceeded"

"show ipv6 interface" does not paginate

block of copper ports 5-8 and 13-16 think they are fiber ports, possible

Shell SSH Remote Access

Instance pull down menu displays 4k

when the maximum number of SSH sessions has been exceeded.

The switch does accept 5 SSH sessions normally, when the 6th session connection is attempted via SSH, the connection is

does not paginate corr ectly when a large number of interfaces/ad dresses are configured.

ports.

Security,

Secure HTTP, Secure Shell does not work

the Web UI, the Instance pull down menu has a range of 1-4094 MSTIs.

A Log message "Unable to find a free connection. Exceeded the maximum number of allowed connections." has been added.

Pagination mechanism corrected.

Corrected error checking logic to correctly select fiber ports for fiber diags and copper ports for copper diags.

Issue is corrected

platforms

negotiation issues with 10GBase-T

Under Port Configuration page get error on 10GB ports when you change any setting

secure ssh from web interface. error 'sshcfg_load start' missing

fails to provide an iSCSI web page when selecting the

6200 and 8024/8024F

example the duplex settings or description, from the GUI under switching -> Ports -> Port Configuration you will get outputs error when changing port settings on 10G ports. The popup says 10G speed is invalid. Error message attached. This is a problem with code ve r s ion 4.1.0.6 under the GUI no

interface. Error pops up.

selected with OpenManage. .

Driver update applied.

Erroneous pop-up message removed.

Corrected web page

Corrected web index p age

platforms

38 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Line "no

The following is displayed when startup-

All 4.2 supported

Web GUI

Switching-->Port-->Port Configuration

PCM6348

Advertise LLDP

LLDP advertises the switch's MAC address

All 4.2 supported

PCM6348 - Log

Messages repeated in t he switch logs

sender MAC is 76:f7:b9:cf:06:7e.

All 4.2 supported

Firmware

the GUI

Port VLAN membership will not be shown

PCM6220

passive-interface Vl32" in startupconfig does not get loaded on startup

config is loaded:

Applying Global configuration, please wait ... Applying Interface configuration, please wait ...

******* The following lines in "startupconfig" failed:

******* Line 57:: no passive-interface Vl32

Corrected stored config

platforms

Admin Duplex setting on 10G Fiber ports

Management Address as IP Address assigned to OOB interface

message output

Select a ten gigabit in terface Te1/0/1 and you can see that "Admin duplex" and "speed" are still active and can be changed, but when you apply it gives an error message saying " Error failed to set current auto negotiation with disable".

as the Management Address. Request to advertise the management IPv4 or IPv6 address instead.

<190> APR 28 19:39: 22 10 . 1.208.10-1 STATSMGR[239558880]: collector.c(1066) 3445 %% Failure in function collector Get <190> APR 28 19:39: 22 10 . 1.208.10-1 STATSMGR[239558880]: presenter.c(102) 3446 %% ERROR!! Failure in preStatsGet 0x0002011e <188> APR 28 16:40: 31 10 . 1.208.10-1 ARP[301140128]: ipmap_arp_api.c(855) 3325 %% Received ARP Reply on interface Vl208 with bad target IP address

255.255.255.255. Sender IP is 10.1.209.60,

Invalid selections disabled on page load.

The following behavior is now implemented:

1) Will advertise service port address if it is there

2) If service port is not th ere, then will send network address

3) If both of the above are not there and if it is routing interface will send IP address of the routing interface

4) If none of the above not there, will send MAC address.

SNMP walk/gets on counters for CPU/vlan routing ports not supported by platform no longer log a warning message

platforms

missing port VLAN membership in

System Firmware Version 5.1.1.7 Page 39

in the GUI.

Web code updated to display VLAN membership

M8024 - 1GB

X520-KX4

The M8024 is negotiating 1GB speed

PCM8024

Feature request -

The "Client-DUID" is used to uniquely

Client-DUID parameter has

All 4.2 supported clock summer-

minutes

Time change offset is applied early.

All 4.2 supported

M6220 - Simple

Modular switch in Simple mode. Interfaces

again.

All 4.2 supported

SSH

after reboot

SSH config is not restored after reboot.

All 4.2 supported

PC8024 -

change

On 4.1.0.6 if VLANS are assigned to a

that are assigned

All 4.2 supported

Web UI displays

After the browser session timeout (default

than 240 minutes

All 4.2 supported

PC8024 - Secure

If you go to system managemen t, Secure

All 4.2 supported Upgrades to

setup interfaces

The wizard in 4.1.0.9 fails to create any or

All 4.2 supported PC6220M GUI

issue didn't exist

The issue using CLI stack ports shows up

All 4.2 supported

link being negotiated on 10GB Intel

Display "clientDUID" in DHCP binding table.

time recurring EU offset 60 zone "GMT" not offsetting the time by 60

mode, VLAN setting not active when move port to another group

connection with the 10GB X520-KX4 Intel card in the blade server. Issue is only present with new

identify an individual host receiving DHCP address configuration. In the case of a multi-homed host, it is possible to deter mine that the host has multiple bindings for different interfaces and which interfaces belong to each host.

have VLAN settings differen t from default. If you move port from one aggregation group to another, VLAN settings are retained and are listed in running config but it is not active until system is rebooted or VLAN information removed and reentered

Incorrect pre-emphasis values have been corrected.

been added to show "ip dhcp binding" command output.

Day of month calculation has been corrected.

None – issue corrected

platforms

configuration not restored

VLAN trunk assignment CLI

incorrect session timeout duration

HTTP Random Characters -

4.1.0.6

4.1.0.9 are causing network outages. The wizard from

4.1.0.9 fails to

trunk, the switch CLI will display the VLANS not assigned instead of the VLANS

10 minutes), Web UI displays incorrect session timeout duration in the message "Your session had been inactive for more

HTTP, you will see ran dom characters populating the fields. Even if you delete these entries and re-populate with valid chars, it will repopulate with invalid characters making it impo s s ible to generate a cert using the web int erface.

very few interfaces and there was no config for any or very few of the ports, thus, the ip, icmp traffic was not working between ports as well.

ssh configuration saved

VLANs are displayed in positive format, i.e. assigned VLANs are displayed.

The displayed message was corrected.

The random character rendering has been co rrected.

Issue with switch upgrade has been resolved.

platforms

shows stackports down but CLI does not. In previous version

40 System Firmware Version 5.1.1.7

and counters shows no transfer rate but GUI shows stack ports down.

Web shows stack ports up if they are up

platforms

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

M6220 -

Customer has multiple M6220 switches,

All 4.2 supported

MIB walk

SNMP MIB walk crashes switch.

All 4.2 supported

M8024 - No

On the web GUI-->system-->management

attachment.

All 4.2 supported

Bug in Simple

Once you select the Tagged -Vlans, there is

only way to unselect those.

All 4.2 supported

M6348 simple

group

Can't remove what running config says is

All 4.2 supported

Banner motd

The "banner motd" config uration changes

All 4.2 supported

Error message

Attempting to assign an ACL to VLAN 1,

All 4.2 supported

Web UI not

Binding an IP ACL as out bound to an

All 4.2 supported

Auto-neg option

Auto-neg is an option in the port

All 4.2 supported

Routing fails on VLAN

and as part of a network change t hey wanted to remove a VLAN rout ing interface that was no longer in use. When they removed the IP address from this VLAN, they lost routing across the entire switch.

Ports are now checked to ensure they are not part of other routing VLANs before clearing the ARP poli cy on the port.

platforms

crashes switch

password min length error using web GUI

Switch GUI.

mode issue with 10G ports in aggregator

configuration changes after reload.

via Web UI refers to "Management VLAN"

security-->password management,

If the password min length is unchecked and applied, there is an entry in the running config for "no passwords min-length". This can be verified through CLI. After this, on enable passwords page if we try to set a password which is any length less than 8, we get an error message as shown in the

NO way to unselect them again via GUI. Using command-line configuration is the

there

after a switch reload.

via Web UI, generates an error message referring to the Managemen t VLAN. "Error! Cannot Bind ACLs to Management VLAN."

A large array is allocated statically instead o f o n the stack.

Password ranges are no w allocated dynamicall y on web pages.

Tagged VLANs can be unchecked in simple mode

Add interface command i s corrected.

Previously, the input mechanism ignored ANY lines beginning with an exclamation point. Now, the CLI engine will allow lines with a leading “!” in the banner

the hard coded value has been

removed in favor of the correct

VLAN val u e

platforms

generating error for out bound ACL & applies the config.

via Web UI is not grayed out for 10Gb fiber ports.

System Firmware Version 5.1.1.7 Page 41

interface, via Web UI, does not generate an error message. Further more, the config gets applied to the interface as an inbound ACL. When the same action is done via CLI, an error message is gener ated and the config is not applied.

configuration via Web UI for 10Gb fiber ports. Ideally the auto-neg option should be grayed out for 10Gb fiber ports.

Use of the direction object has been corrected.

Auto-negotiation selection is disabled for 10G ports in the Web as auto-negotiation must always be enabled.

platforms

CLI command

The CLI command “show interface detail

All 4.2 supported

Creating

config

Using web GUI to create and assign ports

All 4.2 supported

Assigning ACL

via Web UI

When binding an ACL to a VLAN

All 4.2 supported

Incorrect Model

The last revision date has not been revised.

Nov 2003 12:00:0 0

All 4.2 supported

Links do not

Link dependency CLI "show link-

of the members or the depended-on ports.

All 4.2 supported M8024 does not

configuration

Show spanning-tree mst-config not

PCM8024

Need to increase

than 50

Previous limit was 20.

All 4.2 supported

On the 8024F,

The PowerConnect 8024F ports 1 and 2 will

PC8024F

The CLI

value

When executing the command terminal

All 4.2 supported

The Service tag

platforms.

When executing the “show system id” is not

All 4.2 modular platforms.

"show interface detail portchannel1" locks up console session

VLANs, assigning ports using vlan membership detail page causes removal from running-

Priority to a VLAN interface does not get set

numbers and revision date in dellref.my

show status

port-channel1” generates the following errors:

Max number of lines in the scroll buffer reached. Output will be truncated.

via vlan membership detail page causes previous vlan ports and sometimes previous vlan to be removed from running-config.

interface, via Web UI, th e "Ass ign ACL Priority" field does not get set as it would if it were configured via CLI.

This could lead a customer to believe they do not need to update their MIB file. LAST-UPDATED "200311210000Z" -- 21

dependency" shows "Member ports" and "Ports depended on" port numbers along with group id, but does not show the status

The scrolling issue h as been corrected.

VLAN creation addition/removal behavior is corrected.

Web interface has been corrected.

The MIB has been corrected. Administrators are advised to load the new MIBs when updating to this release.

The display has been updated to show the requested information

platforms

save complete MST vlan

the VRRP routing instances or VRID to greater

inserting or removing the cable on ports 1 and 2 causes both ports to shut down briefly

command terminal length is not setting terminal length

is being deleted on modular

working.

flap briefly when cables are r emoved and reinserted.

length <value>, the value is not updated after execution.

showing the service tag on modular platforms.

All MSTP configurations are now saved.

VRRP instances are now set to

50.

All inter-phy operation is fixed.

Terminal length settings now take effect immediatel y.

The service tag was accidentally being deleted and is now being saved correctly.

platforms

42 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Resolution

Affected Platforms

PC7048R-RA switch has boxes

PC7048R-RA switch has cpu process boxes

Corrected CPU utilization

PC7048R-RA

The warning message about IP address

one subnet.

Add warning message in

port

Config commands did not

3.1.5.13

Command "spanning-tree mode mstp"

3.1.5.13.

Auto-neg option via Web UI is

ports.

Auto-neg option via Web UI is not grayed

Simple mode, VLAN setting

moved to another group

if a port is moved from one aggregation

retained but not applied until reboot.

Unable to configure secu re SSH from web interface.

Error message 'sshcfg_load start' missing is returned.

Secure HTTP Random

Random characters populate the Secure

IP PIMSM BSR/RP Mapping

When the RP or the BSR changes, the data

get software forwarded.

Error messages when issuing a

Accessing unsupported counters causes

Release 4.1.1.9

req cpu process 38-54% utilization with single power supply.

Router crashes on OSPF network type change.

Occasional crash when configuring VRRP.

Cannot create Dynamic LAG with Interface range co mmand

PC8024 cannot forward packets on port9 to port16 while linked on at 100Mb.

Show fiber-por ts opt ic a l transceiver is not displaying the correct interface numbers.

VRRP routing instances increased to 50.

DHCP on in-band and out of band ports

req running betw een 38-54 % utilizat ion if it’s running either only with primary or secondary power supply.

The switch can crash when changing the configuration from the default of broadcast to point-to-point.

With routing globally disabled, bouncing VRRP on a host interface occasi onally causes a crash.

Dynamic LAG cannot b e cr eated from CLI using interface range command.

PC8024 cannot forward packets on port9 to port16 while linked on at 100Mb.

Instead of reporting on exact ports that had modules, diagnostics were reported on the ports that did not have modules

VRRP routing instances increased to 50. Increased VRRP routing

conflict is not printed in case DHCP is configured on in-band and out-band ports and they both receive an IP address from

issue when running with one power supply.

Corrected process synchronizati o n pr o bl e m

Correct the transition between routing being disabled and routing being enabled.

Correct the command syntax to "auto " instead of "active"

Correct interoperability problem between the PHY and the switching core.

Corrected register pr oblem when accessing the SP F+ ports

instances table.

case offered IP addres s is conflicting with the configured one on another

All Platforms

PC8024 PC8024F

PCM8024

All Platforms

migrate correctly from vers ion

not grayed out for 10Gb fiber

not active when a port is

Characters

is not robust

show statistics command while traffic running.

System Firmware Version 5.1.1.7 Page 43

doesn't migrate correctly from version

out for 10Gb fiber ports for the PCM8024-

group to other, VLAN settings are

HTTP web page.

traffic may get affected and in some case

error messages to be displayed.

Correct the command tree to migrate the old syntax.

Corrected the Web UI. PCM8024-k

Correct initialization when port is moved.

Corrected Web page error . All Platforms

Corrected Web page data initialization.

Corrected RP join processing.

Change the logging priority of the messages for unsupported counters so that they are not output.

All Platforms

PCM6220

All Platforms

Summary

User Impact

Resolution

Affected

When network re-configuration happens,

traffic.

Port Configuration page gets

any setting is changed.

IGMPv3 querier sent in error

DNS client error in logs and

Message: “nimCheckIfNumber: internal

"no enable authentication"

telnet line

If LAG port is configured as general trunk

agentLagSummaryAccessVlanID.

PCM6348 - Log message output

Missing port VLAN membership in the GUI.

Routing fails on VLAN. Removing a VLAN routing interface

PIM BSR join messages. PIM Join messages using wrong RP

PIM-SM Not al l data passed down to the RPT.

SNMP walk through unsupported counters cause error messages

If switch has less than 24 ports, Port VLAN membership will not be shown in the GUI.

causes routing to b e lo s t across the entire switch.

Address

the multicast traffic may not properly converge thus resulting in loss of some

Platforms

Change the logging priority of the messages for unsupported counters so that they are not output.

Corrected the VLAN port data issue.

Corrected the VLAN interface removal problem

Corrected Join without a Prune condition.

Corrected Join condition. All platforms

All Platforms

PC70xx

PCM6220

All Platforms

errors on 10GB ports when

PIM-SM RP Fails to send Register Stops.

with source address 0.0.0.0

PIM-SM Jo ining messages using wrong RP address

Fails to reconfigure and forward multicast messa ge

switch locks up and needs to be rebooted.

PIM-DM Prune states expiring.

Email alerts do not contain log message.

log message "internal interface number 183 out of range" is seen.

does not disable access to privileged EXEC over the

Port Configuration page gets errors on 10GB ports when any setting is changed.

Switch will get moved to non-querier mode if an IGMP membership query arrives with source address 0.0.0.0.

PIM-SM Joining messages using wrong RP address

Switch fails to reconfigure and forward multicast messages following link failure

Error message “DNS Client: osapiSocketRecvFrom returned error for address 0xAABBCCDD” being logged

The router might end up sending Graft messages even thou gh there are no intended hosts.

Current e-mail messages display in e-mail applications with a blank body, log message is not displayed.

interface number 183 out of range” is displayed.

"no enable authentication" does not disable access to privileged EXEC over the telnet line

Corrected Web page error . All Platforms

Corrected Register Stops message error.

Corrected valid address check for membership query packet.

Corrected Join message processing.

Correct an RPF information problem when link goes down.

Corrected error in tas k t hat handles changes of the Service Port link status

Corrected a PIM-DM timeout error.

Corrected error in email header section.

Corrected error in VLAN create web page

Corrected error in authenticati on pr oc e s s ing.

All Platforms

dot1q_api.c error during MIB walk

sFlow Timeout settings for CLI command

44 System Firmware Version 5.1.1.7

then log message is getting printed when snmpwalk is accessing the MIB OID

sFlow receiver timeout value should be optional

Corrected VLAN access error.

Changed receiver timeout value to be optional.

All Platforms

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

The sFlow agent field isn't being filled into

Stack module inserted in Bay

Summary User Impact Resolution Affected

Platforms

sFlow: Limit of 32 for the number of interfaces for which sFlow can be enabled

sFlow: Formatting errors with some of the records

sFlow: The sample Pool field is not being incremented properly

sFlow: The sFlow agent field isn't being filled in

1 and SFP+ module in Bay 2

CLI "switchport gen eral acceptable-frame-type" different options than GUI

startup-config cannot be restored via tftp

"show ip ospf database" does not respect configuration of "terminal length 0"

Occasionally, "clear con fig" from console triggers wio_api.c traceback

Radius config stops working after reboot.

ISDP timers running fast. ISDP entries are timing out and dropping

VRRP hang VRRP los s of mastership causes the switch

IPMC error messages and forwarding issue.

Dynamic LAG Ports are flapping/continuous trace messages.

Simple Switch should have FC enabled by defa ul t

Need to change message on console when try to downgrade anything below

4.1.0.19 release for new DDR

Current limit of 32 for the number of interfaces for which sFlow can be enabled.

Some of the sFlow records had formatting errors.

Sample pool field will not be incremented in the sFlow packet sent t o the collector.

sFlow Packet. With the presence of stack module at bay1,

SFP+ module in bay2 will not get detected until the stack-module at bay1 is removed.

The CLI "switchport general acceptableframe-type" contains different options than GUI.

TFTP script download or saved script validation will leave the DUT in an unusable state.

The CLI handler of the "show ip ospf database" command does not consider the terminal length configuration.

Occasionally, "clear con fig" from console triggers wio_api.c t raceback

Radius config stops working after reboot, needs to be removed and re-added in order to work again.

off the list before the next update comes in.

to hang.

Error messages indicating that IPMC table are not set correctly in hardware.

Lags do not stay up on combo ports. Corrected an error in

Enable Flowcontrol on simple mode. Flowcontrol enabled on

If a switch contains new DDR, downgrading to 4.1.0.19 is not allowed.

Changed the limit to allow all sFlow instances.

Corrected sFlow header problems.

Corrected sample pool field error.

Corrected problem adding sFlow agent address.

Corrected error that prevented the detecting of SFP+ module in bay 2.

Options are now the same. All Platforms

Corrected script validation problem.

Corrected terminal len gt h configurati on ha nd l ing.

Corrected “clear config” internal message error.

Corrected error in Radius user authenticatio n processing.

Corrected an error in the ISDP timer disable processing.

Corrected an error in the VRRP processing of the new IP address.

Corrected error message processing which generated false error messages.

mapping the physical ports to a LACP logical port.

simple mode. Modified the downgrade

message.

All Platforms

PCM6220

All Platforms

PCM8024 PCM8024-k (with new

DDR)

System Firmware Version 5.1.1.7 Page 45

Release 4.1.0.19

Summary

User Impact

Resolution

Affected

PC8024F/ All Platforms

All Platforms PCM8024-k

Combo ports flapping when fiber port connected to certain other vendor switches

New M8024/M8024-k cards have different memory parts requiring changes to DDR timings

10GBaseT ports do not come up

Switch crashes when executing “show interfaces detail te1/2/1” command with no space between “te” and the unit/slot/port identifier

OOB Default Gateway does not save when using CLI setup wizard

Upgrades to 4.1.0.6 are causing network outages. The wizard from 4.1.0.6 fails to setup interfaces

The combo port is not coming up after performing repeated plug-out and plug-in.

Line "no passive-interface Vl32" in startup-config does not get loaded on startup.

Unable to configure Secure Shell from web interface.

OpenManage web interface fails to provide an iSCSI web page when selecting the iSCSI item.

Wrong information for current link status in port configurati on pa g e .

PFC sends packet after quanta extension received.

CLI command "show ipv6 interface" does not paginate correctly.

MOTD Banner appears at t he wrong time.

Platforms

Users unable to reliably operate over combo ports

User unable to downgrade t o earlier version of firmware

Unable to use 10GBaseT ports on PC7048R-RA

Service inter rupt ion due to crash Issue has been corrected. M8024

There was no way to configure the gateway on Out-Of-Band interface.

Ports on some devices will not attach. Corrected the Configuration

The fibers port associated with Fiber / RJ45 combo ports may flap or not link up with certain other switches.

The "no passive-interface Vl32" configuration does not get loaded from the startup-config on startup.

Proper web page operation is prevented. Web page has been

The web page for iSCSI fails to open when selected with OpenManage.

Incorrect stacking link status is presented in the Web page.

This can cause FCOE failures when using PCM8024-k as a tran s it switch between some switches.

A large number of interfaces cause information to scroll off the viewing area.

MOTD and Login Banner is not visible before login for SSH users.

Corrected combo port flapping issue with new PHY firmware. Added new media-type command to configure the preference on a combo port

No resolution. Newer IOMs do not support earlier versions of firmware.

Corrected internal addressing of port. Ports now come up properly.

Add functionality for configuring gateway on Out-Of-Band interface.

Wizard to properly apply the configuration to all ports.

Add commands to allow operator to pref er or force RJ45 or SFP port selection on combo ports.

Corrected the loading of the "no passive-interface Vl32" configuration.

corrected. Web page has been

corrected.

Web page output has been corrected.

Switch no longer sends packet after quanta extension received.

Fixed the pagination for the command.

Fixed the banner proc e s s i ng function.

PC8024 PC8024F

PCM8024 PCM8024-k

PC7048R-RA

M6220 M6348 M8024-k

All Platforms

PC8024

All Platforms

PCM6220

All Platforms

PCM6220

46 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

PCM8024

All Platforms

PC8024

Model numbers which are Corrected VLAN instan ce Correct the handling of

Corrected the ability to

the running-config.

Corrected the port mapping

configuration.

PCM8024

Added an error message.

All Platforms

Corrected the fetch o f

All Platforms

Correct password length

All Platforms Corrected a memory

All Platforms

Corrected the Web page t hat

PCM8024

Summary User Impact Resolution Affected

Platforms

Summertime function running early.

Switch does not save complet e MST VLAN configuration.

Switch failed to generate new RSA/DSA crypto keys.

Banner MOTD configuration changes after reload.

VLANs can get removed from running-config.

Console session locks up. CLI command "show interface detail port-

Fiber diagnostics reporting incorrect port type.

BROADCOM-REF-MIB reporting incorrect values.

Switch crashes when full vlan range assigned to MST regions.

Config Wizard causes iSCSI error messages.

SSH configuration not restored after reboot.

The summertime function resets the time on the wrong week.

The correct configuration was not being saved if there are more than 1K VLANs in a single MST instance.

“Failed to generate RSA key.” message was produced and keys were not generated.

After reload a MOTD with an embedded! Character will be truncated .

Creating vlans, assigning ports using the vlan membership detail page causes removal from running-config.

channel1" locks up console session.

When running fibe r diag nostics, Ports 5, 7, 8, 13, 15, 16 report to be fiber ports and not copper ports.

Incorrect Model numbers and revision date in BROADCOM-REF-MIB dellref.my

When the full range of VLAN instances is assigned to a MST region the switch crashes.

Using the config wizard to setup the PC8024 causes iSCSI error messages.

The 'ip ssh protocol 2' is not being included in the running configuration.

Corrected the function that returns the day of the week to account for parti cul ar months and leap years.

Corrected the VLAN configurati on buffer size.

Fixed error condition when generating keys.

Fix the MOTD engine to not ignore the leading ‘!’ character.

Corrected default VLAN checks when updating VLAN information.

Corrected buffer overflow during "show interfaces detail" execution.

Fixed reporting of valid copper ports.

described in dellref.my are corrected.

buffer overflow.

iSCSI admin mode.

place 'ip ssh protocol 2' in

All Platforms

PC8024 PCM8024

PCM8024

All Platforms

PC70XX

PCM6220

PC8024

All Platforms

1GB link being negotiated on 10GB Intel X520-KX4

No message logged for "Max number of SSH login sessions exceeded"

VLAN web page not updating. When VLANs are removed using the web,

No password min length error when using web GUI.

MIB walk crashes switch. When running a large SNMP MIB walk

Unable to deselect VLAN tags in web GUI.

System Firmware Version 5.1.1.7 Page 47

Some internal links on the Modular switches will not link at 10Gb.

When the Max number of SSH sessions is reached no error is logged.

the GUI doesn’t display the ports that are members of the VLAN.

If "no passwords min-length" Is set an error results if the password is less than 8 characters.

the switch crashes. Once Tagged-Vlans are selected there is

no way to unselect them again via the web GUI.

to allow proper port

current VLAN data.

range check.

problem in the MIB walk.

allows tagged VLANs to be unselected.

PCM8024-k

Summary

User Impact

Resolution

Affected

Access is allowed to files on

Corrected the web process

All Platforms

Corrected download web

All Platforms Corrected task table

All Platforms

Corrected task table

All Platforms

Corrected the code looping

All Platforms

Corrected the code looping

All Platforms Correct the SDRAM test

PCM8024-k

The help page for the PCM8024-k

Corrected the help page.

PCM8024-k

When the switch boots, If the CRC check of the VPD fails, the switch will hang.

Corrected check for valid VPD before CRC check.

All Platforms

Improved Performance with High

Improved performance.

PC70XX

PC6348

PCM6220 Web pages did not have the

Updated the PCM6220 web

PCM6220

Summary

User Impact

Resolution

Affected

the switch without log in permissions required

Switch crashes when downloading without the .stk extension.

Switch crashing when HTTPS session is enabled.

Switch crashing when trying to open more than 3 HTTPS sessions

Mozilla Java Script er r or when selecting VLANs.

IE7 error when selecting VLANs.

SDRAM test does not run the specified number of iterations.

PCM8024-k Web page an d help page mismatch.

VPD CRC check hang

If the file name is known, th e fil e can be downloaded through the web browser without having to be logged in.

Switch crashes when trying to download code without .stk extension through HTTPS using HTTP.

Switch is crashing when HTTPS session is enabled from CLI but not from WEB

Opening more than 3 or more HTTPS sessions will crash the switch.

In Mozilla, when 1-4085 VLANs are *selected* in the web page for Tagged VLAN Java Script error appears.

In IE7, when 1-4085 VLANs are *selected* in the web page for Tagged VLAN IE7 error appears.

When running the SDRAM test a <cntl-c> must be used to exit the test.

Switching--> Network Security--> Dot1x Authentication--> Authentication Detail page did not match the page.

Platforms

to require login credentials before downloading a file.

page.

memory size.

error.

iteration loop.

Packet Buffer Optimization.

PCM6220 Stack View needs PCM70XX styling

Release 4.1.0.6

SSH crash - memPartAlloc: block too big

PC M8024 switch reset out-ofband address to none when switchports were changed

Web page shows IP address as '0.0.0.0' for '1.1.1.1' routing interface.

Read-Only Web page is populating all configured IP and IPv6 ACL names when we select the ACL Name.

Utilization iSCSI Workloads.

same styling as PC70xx.

Reduced switch functionality. Memory allocation issue is

Inability to access switches via OOB port. The out-of-band address is

Potential operator confusion over switch operations

Potential operator confusion regarding web page operations.

pages.

corrected and checked fo r memory leaks

maintained over switchport changes.

The web page output has been corrected.

The web page has been corrected to only populate the selected entry.

PCM8024 PCM8024-k PC8024

Platforms

All Platforms

PCM8024

All Platforms

48 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary User Impact Resolution Affected

Platforms

FAN LED graphic on web page needs to glow in RED when FANs are not operational (stopped).

Incorrect command is being displayed in running-config, when boot host dhcp is disabled.

DUT crashes while configuring max dynamic VLANs.

Manager of the stack is changing when trying to learn maximum number of VLANs using GVRP.

Switch prompts to save config data when no changes have been made

Crash while RFC3918 Group Capacity test is running

'no' version of 'key' command is not implemented

Password is not accepting quotation ( " ) character

Incorrect warning message displayed while executing the command "boot system <unit> image1"

IPV6 command displays wrong output

LLDP-MED log messages showing 5 sec difference in entry age out information

DHCPv6 web issues Potential operator confusion regarding

LLDP MED application should not allow configuration of location and inventory transmit TLV's as underlying application not pr e s e nt

Inability to determine switch status. Web page has been corrected All

Platforms

Potential operator confusion regarding web page operations.

Network outage possible. The PowerConnect does not

Potential operator confusion regarding switch operations.

Potential operator confusion regarding switch operation.

Network outage possible. The PowerConnect switch

Potential operator frustration with switch management.

Potential operator confusion regarding switch configuration.

Potential operator confusion regarding switch operation.

switch configuration.

Potential operator confusion regarding switch configuration.

The running config now shows the correct configuration

crash when using maximum dynamic VLANs

The stack manager does not change during learning with GVRP

The switch no longer prompts to save config data if no changes have been made

runs the RFC3918 test without crashing

The no key comman d is implemented to return the key configuration to the default.

Passwords can be enclosed in quotes (contain embedded blanks). A password may not contain a quote. The accepted character set and length is documented in the CLI reference manual.

The error message has been corrected to indicate that the unit selected for reboot does not exist.

The IPv6 output has be e n corrected to remove the duplicate display lines in show ipv6 he lp.

The LLDP timer has been updated to account for processing skew.

The acceptable ch ar acter sets are documented in the CLI Reference guide.

Location and inventory TLVs cannot be enabled for transmission in LLDP MED.

All Platforms

System Firmware Version 5.1.1.7 Page 49

Summary

User Impact

Resolution

Affected

SysUpTime is not being shown correctly duri ng an SNMP walk (Poll interval 1sec)

Log messages need to be corrected on ip dhcp snooping rate limit scenario.

The show ip vlan command output is not proper after morequit prompt is encountered, i.e. after around 16 routing interfaces

Invalid error port number displayed on log message when vlan is changed to forbidden mode from access mode

The banner motd XXXXX does not appear in show running-config

Web page mac-vlan table too slow to load.

Confused between ip default gateway and ip default route (update manual with how to set a default route).

Auto Install show boot retry count line needs to be left aligned by one space.

Cannot Access Optical Transceiver Diagnostics Page if Multiple Submits done prior to initial refresh completing

Error messages for non-existent stack members non-informative

Traffic is forwarding when IPv6 forwarding is disabled

A LAG member comes UP if configured individually as no shutdown, even though the port-channel's state is d own

Potential operator confusion regarding switch operation.

Potential operator confusion regarding switch configuration. Inability to diagnose network issues.

Potential operator confusion regarding switch configuration.

Potential operator frustration with switch management.

Potential operator confusion regarding switch configuration.

None. The retry count alignment is

Potential operator frustration with switch management.

Potential operator confusion regarding switch configuration.

Incorrect operatio nal state in network. The ipv6 forwarding

Incorrect operatio nal state in network. LAG members are placed in

The correct variable is u sed to write SysUpTime

Interface representations in log messages use unit-slotport format.

The paging has been corrected.

The error message is no longer issued.

All banner configuration appear in the running-config

For certain browsers, paging has been implemented t o speed up load times.

The ip default route command is deprecated. Use the ip default route command to set a default route.

corrected

This is a browser dependent issue (IE 6) that is not seen in later versions. Th e web session recovers after d oing a refresh.

The user can pre-configure stack units. If the stack unit does not exist for a switch configurati on ope r a ti o n, a n error message indicat ing same is issued.

command is deprecated. To disable traff ic forwarding, use the “no ipv6 unicast-routing” in place of the “no ipv6 forwarding” command.

the blocking state for dynamic LAGs and only come up when the LAG link comes up.

Platforms

All Platforms

PCM8024

All Platforms

PC70XX

PCM8024

All Platforms

50 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary User Impact Resolution Affected

Platforms

Re-authenticate Now check box is not highlighted when edit check box is selected

Block command is not seen in show running config through web and cli

FDB entries are getting aged out before default age-out time, when both FDB and MFDB tables are full.

DHCP packets forwarding i s not proper to/from the trusted and un-truste d port s

Configuring all ports in all VLANs takes a long time

Max number of OSPF neighbors not su pp or te d

Cable fault distance not getting displayed in WEB in case of cable with one or more pairs cut or short.

Ports representation need to be changed in debug messages

No error message for illegal characters in various command parameters

The ARP entry is not seen in the ARP table when an ARP reply is sent to DUT.

Mapping Table configuration is not being displayed on ReadOnly mode user web page.

ACL is getting deleted when trying to create max+1 rul e.

LLDP-Med TLV information not registered for jumbo frame sizes greater than 8000

VLAN binding entries are not being displayed on Read-Only mode user web page.

Inability to configure switch. The Re-authenticate Now

check box is highlighted when the edit check box is selected

Potential operator confusion over switch operations.

Incorrect operatio nal state in network. The frame flooding routing

Potential operator frustration with switch management.

Potential operator confusion over switch operations.

Potential operator confusion over switch operations. Inability to diagnose network issues.

Potential operator confusion over switch configuration.

Potential operator confusion over switch operations. Inability to diagnose network issues.

Potential operator confusion over switch configuration.

Potential operator confusion over switch operations. Possible security issues.

Potential operator confusion over switch operations.

The block command is a temporary administrative assignment and is not maintained persistently in the saved or running configs.

On the PCM6220, the MFDB and FDB tables are a shared resource. The user is continually sending new MFDB entries, which causes old FDB entries to be removed to make a place for the new MFDB entries.

in DHCP snooping now takes into account trust status.

VLAN configuration has been optimized.

The maximum number of neighbors is supported.

The fault distance is displayed in the web page.

Ports are now displayed in C/S/P standard format

The accepted character set and length is documented in the CLI reference manual.

ARP entries are stored properly

The mapping table is displayed for a read-only user.

Corrected error checking logic.

Jumbo LLDP frames are now processed properly

VLAN binding entries are available to read only users

All Platforms

System Firmware Version 5.1.1.7 Page 51

Summary

User Impact

Resolution

Affected

Unable to execute the command dot1x timeout txperiod 1

In switching> network security >dot1x authentication web page in read-only user mode, the Re-Authenticate Now check box can be checked

System Device information web page LED information not in sync with front panel LED information.

DHCP snooping static binding thru DHCP request is denied as this is an expected behavior.

Mode of transfer is displaying “unknown,” while downloading the code from ftp

Allow disabling and enabling of terminal paging

Web does not allow to configure image descriptor to its max length i.e. 255 characters.

CDP (ISDP) is a c tive on portchannels instead of the member Ethernet interfaces. For dynamic LAGS, the ISDP information is not exchanged on the interface until the portchannel becomes active.

CLI Manual Has No Index Potential operator frustration over switch

Script validation is fails when max SNTP servers are configured.

Configured SYSLOG server parameters cannot b e updated without deleting and reconfiguring

Potential operator confusion over switch operations.

Potential operator confusion over switch operations. Possible security issue.

No Potential operator impact expected. This issue was regarding

Potential operator confusion over switch operations.

Potential operator frustration over switch configuration.

Potential operator confusion over switch operations.

Potential operator confusion over switch operations. Inability to interoperate with other switches.

configuration. Potential operator confusion over switch

configuration.

Potential operator frustration over switch configuration.

This command is accepted with a timeout period of 1 second. Corrected range check on input.

Read-only properties are s et for the check box

various stylistic aspects of the system device web page. The system device web page conforms to the requirements as it exists and does necessarily match th e C LI with regarding to capitalization or naming conventions

Removed the log message indicating that a bound DHCP client with an existing binding sent a DISCOVER. Added a counter for this condition to the DHCP debug statistics.

ftp transfer mode is displayed

The terminal length command is now implemented.

Images descriptors up to 255 characters are allowed

CDP is active on the member ports for dynamic LAGs when the LAG is active.

The CLI manual has an index.

Corrected CLI validation check so that existing ser ver can be entered multiple times.

Syslog server parameter can be updated in the CLI without deleting the server.

Platforms

All Platforms

52 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

LLDP Assignment of port ID for

Management ACL list needs

SNMP support for D ell-LAN-

Simple mode BPDU w ill pass

single link per aggregator

Fatal error cra sh and reload when

Interface metric not working on Local system MAC address

channel"

SNMP going unresponsive on

Network perf ormance and

H323 VoIP traffic and

LACP not working with Juniper

PDUs.

Summary User Impact Resolution Affected

Platforms

CLI will not let user configure available parameter for the given IGMP command

Inconsistent b ehavior - using same port number for multiple services

Port-Descrip tion TLV

"Match every packet" option

TRAP-MIB

thru port-aggregator to other port-aggregator when using a

Potential operator frustration over switch configuration.

Potential operator confusion over switch operations.

Potential operator confusion over switch operations. Inability to diagnose network issues.

Potential operator frustration over switch configuration. Possible security issue.

Potential operator frustration over switch configuration.

Network outage if conn ected to Cisco with BPDU-Guard enabled

IGMP configuration commands can be enter ed in interface VLAN mode at any time.

Attempting to add a service with a TCP port overlapping a TCP port used by an existing service is denied with an appropriate error message.

The LLDP port id TLV is supported by Powe r C onne c t and can be displayed on peer devices.

New syntax has been added to the management ACL to allow the any specification for the service type.

The Dell-LAN-TRAP-MIB is supported.

BPDU flooding is now disabled on aggregator ports.

All Platforms

uplinked to Cisco 6509 - 3.1.3.9 Believe to be ISDP

8024 OSPF passi ve int erface

displayed as partner MAC address in "show lacp port-

modular switches within 6 hours

connectivity is sues. Evidenced by “l7utils_inet_addr” messages in log.

Crashing/Rebooting switch. CPU PID '17c31570 voipTask' spikes as seen in the 'show p rocess cpu'.

EX-4200, show inactive port channel state. System is sending oversize (132 bytes) LACP

Inability to connect to Cisco 6509 CDP enabled ports

Interface metric is adverti s ed as 0 when configured as passive – passive interface preferred over active in terface

Potential operator confusion over network topology

Potential operator frustration over switch operations

Potential operator frustration over switch operations. Switch degradation in VoIP deployments.

Potential operator frustration over switch operations. Inability to interoperate with Juniper switches.

The switch now handles CDP mega-packets of size

1376. The interface metric is

advertised properly when the interfaces is configured in passive mode.

The partner MAC address i s displayed in the Partner system MAC address field.

SNMP operations have been corrected and tested without issue for over 24 hours.

The internal timer operations have been highly optimized resulting in lower overall CPU utilization.

The VOIP QOS code is no longer processing VoIP packets when it is admin disabled.

The LACP process now sends 128 byte LACP PDUs. Interoperability with Juniper switches is verified.

PCM6220

All Platforms

System Firmware Version 5.1.1.7 Page 53

Summary

User Impact

Resolution

Affected

Stack splits into multiple masters

when MU fails.

VLAN name issues in web UI on

switching to new VL A N.

Discards Occ urring After STP topology changes when an

ARP broadcast traffic not

modular. Old VLAN name is retained on web page when

Potential operator frustration over switch operations.

Potential operator confusion over switch operations.

Issue in SDK MAC reset function resolved such that connected unit respond to master within 1 ms. Stacking stability verified under the conditions where issue occurred.

The correct VLAN name is always displayed.

Platforms

All Platforms

Firmware Update to current version 3.1.5 .4.

OOB SSH connection is established

forwarded through static/dynamic 10Gb LAGs

Potential operator confusion over switch operations.

Potential operator frustration over switch operations.

Potential operator frustration over switch operations. Potential network instability.

The wrong counter was being displayed in the CLI. The CLI commands “sho w statistics port-channel” and “show statistics <inter f ace>” are corrected to use the s ame counter as SNMP.

The SSHD task was starving other tasks in the system and causing topology changes. The SSHD task priority is now set to the shared time slice priority.

A field was being utilized in the LAG hash algorithm that could result in different hash indices on different units of a stack. This field (modid) has been removed from the hash selection algorithm.

All Platforms

54 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

ip igmp snooping querier query-interval

The guideline under this command refers to the IGMP Snooping Max Response

Max Response Time is 25 sec

terminal monitor

The command does not have any arguments or a particular expected output hence

( in privileged exec mode to display logging messages on terminal )

ip mtu

The guideline should be augmented with the below text:

outgoing interface.

rate-limit cpu direction input pps

no rate-limit cpu direction input pps

Use the rate-limit command to reduce the amount of unknown unicast/multicast

Note: This command is not available on the PCM6220 switch.

show system internal pktmgr internal control sw-rate-limit

The show command is to display the configured CPU rate limit for unknown packets

Note: This command is not available on the PCM6220 switch.

CLI Reference Guide Updates

Release 5.1.1.7

 No Updates

Release 5.1.0.1

The following table list s issues found in the CLI Reference Guide after publication:

Command Issue

pps_value

Time. A reference that this can be configured using command ip igmp query-maxresponse-time under IGMP commands is missing. The default IGMP/MLD snooping

an example for this is missing in the guide.

prompt#: terminal monitor

Use the ip mtu command in Interface Configuration mode to set the IP Maximum Transmission Unit (MTU) on a routing interface. The IP MTU is the size of the largest IP packet that can be transmitted on the interface without fragmentation. Packets that are L2 forwarded in hardware ignore the IP MTU. Packets that are L3 (VLAN routing) forwarded in hardware observe the IP MTU. Packets forwarded in software are dropped if they exceed the IP MTU of the

packets forwarded to the CPU. Use the “no” form of the command to set the rate limit to the default value.

pps_value – range 100-1024 packets per second (100-3000 for PC81xx switches). The default ingress rate limit is 1024 packets per second (3000 for PC81xx switches). This command allows the administrator to reduce the rate limit for which unknown un i cast and multicast packets are forwarded and/or copied to the CPU. It does not affect the rate limits for control plane packets. It is almost never necessary to use this command to change from the default value. The use of this command should be restricted to situations in which moderate to high rates of unknown unicast/multicast are contin ually sent to the switch CPU as evidenced by the show proc cpu command and where the ipMapForwardingTask is showing high CPU usage. This will occur most frequently in networks where a high number of ARPs are continually received on untrusted ports, high numbers of L2 stations are timing out and reappearing or multicast flooding is occurring in the network. If problems with L2, L3 or multicast learning occur after changing this value, set the rate limit back to the default value and take other steps to correct or mitigate the underlying network issue directly.

in packets per second.

System Firmware Version 5.1.1.7 Page 55

Release 5.0.1.3

Show snmp filters Snmp-server filter

The following note should be added to the command usage guidelines.

aaa authorization {exec } {default | <list_name>} method1[method2]

no aaa authorization {exec} { default|<list_name>}

authorization exec [default |<list_name>]

no authorization exec

debug aaa authorization exec

no debug aaa authorization exec

The following Exec Authorization CLI commands are missing from the CLI reference

Command

Issue

The VLAN name command has been migrated to VLAN Config mode

 No Updates

Release 5.0.0.4

Command Issue

Release 4.2.2.3

 No Updates

Release 4.2.1.3

When a filter is defined, SNMP treats the filter as having an 'exclude all' statement at the beginning of the filter. Unless an include statement is specified, all notifications will be excluded.

guide

aaa authorization {exec } {default | <list_name>} method1[method2] no aaa authorization {exec} { default|<list_name>} authorization exec [default |<list_name>] no authorization exec debug aaa authorization exec no debug aaa authorization exec

Storm-control bro a dca s t

name "RDU-NOC Management VLAN "

Release 4.2.0.4

 Please refer Dell PowerConnect CLI Reference Guide for more details

Release 4.1.1.9

 No Updates

56 System Firmware Version 5.1.1.7

The supported syntax is Storm-contr ol broadcast [level rate] where rate is a parameter to level and defined to be:

The storm-control threshold as percent of port speed. Percent of port speed is converted to PacketsPerSecond based on 512 byte average packet size and applied to HW.

If level is not used the default rate is 5.

and is no longer available in interface VLAN mode.

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Release 4.1.0.19

The following commands are supported on switches that have combo ports:

Comma n d me di a -type

Syntax:

media-type { auto-select [rj45 | sfp ] | rj45 | sfp }

auto-select rj45 - utilize RJ45 media when both media types are active auto-select sfp - uti lize the SFP media when both medi a types are active auto-select – return the selection to the default (auto-select sfp) rj45 – force connection on the RJ45 port. Power off SFP media port sfp – force connection on the SFP port. Power off RJ45 media port

Default Configurati o n:

The default is “media-type auto-select sfp”

Command Mode:

Interface Config mode

Description:

Select the media type for the interface. This command is only valid on combo ports.

User Guidelines:

When both media types are co nnected, the preference as determined by the auto-select keyword parameter selects the active media. When the au to-select keyword is no t specified, the selected media type is powered on and th e alternate media type is powered off. Note that when the auto-select keyword is used with any media type, the SFP port will remain powered and the laser, if any, will remain on in order to allow connections over the SFP port.

Examples:

! Select the RJ45 port and power off the SFP port console(config-if-Te1/0/24)#media-type rj45 ! Prefer the RJ45 port and leave the SFP port powered on console(config-if-Te1/0/24)#media-type auto-select rj45

Command show interfaces media-type

Syntax:

show interfaces media-type

Default Configurati o n:

N/A

Description:

Display the configured and active media type for the combo ports

User Guidelines:

N/A

System Firmware Version 5.1.1.7 Page 57

Command

Issue

show service-policy {in|out}

Examples:

console#show interfaces media-type

Port Configured Media-Type(s) Active

--------- --------------------------- -----Te1/0/21 auto-select, SFP preferred SFP Te1/0/22 auto-select, SFP preferred SFP Te1/0/23 auto-select, SFP preferred Down Te1/0/24 auto-select, SFP preferred Down

Release 4.1.0.6

The Dell PowerConnect CLI Reference Guide is completely new. Users are re ferred to the Dell PowerConnect Configuration Migration White Paper for information on how to migrate configurations from previous releases of Dell PowerConnect firmware to the 4.0.0.6 Dell PowerConnect firmware.

The following table lists issues found in the CLI Reference Guide after publication:

show service-policy in show copper-ports cabl e -

length

The supported syntax is This command is depr ecated. Use the show copper-ports tdr

command to display the stored information regarding cable lengths and the test copper-port tdr command to perform a cable length test. Testing a port brings the port down momentarily.

58 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

The following paragraphs need to be added to the User’s Guides for all platforms except the PCM6220:

numbers of L2 stations are timing out and reappearing or multicast flooding is occurring in the network. If problems with

and its use.

Issue

The following message needs to be added to warn the user that Hotplug of a module is not supported if one of the ports

The following copper SFP needs to be listed as the supported module. However, diagnostics are not supported on copper

Finisar FCLF-8521-3

When a filter is defined, SNMP treats the filter as having an 'exclude all' statement at the beginning of the filter. Unless an include statement is specified, all notifications will be excluded.

User’s Configuration Guide Updates

The following table lists issues found in the User’s Configuration Guide after publication:

Release 5.1.1.7

 No Updates

Release 5.1.0.1

Issue

Unknown unicast and multicast packets are copied to the CPU on the lowest priority QoS queue. Unknown packets are those that do not have hardware forwarding entries. Known uni cas t/mul ticast packets are hardware forwarded and are not queued to the CPU. Control plane packets (e.g. spanning tree BPDUs) are copied or forwarded to the CPU on higher priority queues. The rate limiting for unknown packets occurs on the internal CPU port and does not affect hardware based traffic routing/forwarding in any way. Typical ly, the switch will examine the received packets in software to check if there is a forwarding en try, create a forwarding entry (e.g., add a L2 MAC address or ARP response), and then either discard the packet or software forward the packet (only occurs during the brief transitional period when the system is actively adding a hardware forwarding entry but the hardware is not yet updated). Processing delays for higher priority packets may occur when the internal CPU queue is continually kept busy handling low priority packets.

A command was created to allow the administrator to reduce the rate limit for which unknown unicast and multicast packets are forwarded and/or copied to the CPU. It does not affect the rate limits for control plane packets. It is almost never necessary to use this command to change from the default value. The use of this command should be restricted to situations in which moderate to high rates of unknown unicast/multicast are continually sent to the switch CPU as evidenced by the show proc cpu command and where the ipMapForwardingTask is showing high CPU usage. This will occur most frequently in networks where a high number of ARPs are continually received on untrusted ports, high

L2, L3 or multicast learning occur after changing this value, set the rate limit back to the default value and take other steps to correct or mitigate the underlying network issue directly.

See the CLI Reference Guide updates section of this document for the description of the “rate-limit cpu” CLI command

Release 5.0.1.3

 No Updates

Release 5.0.0.4

on the module is operating in stacking mode.

Warning: One of the modular ports is operating in stacking mode. Hotplug of a module requires system reboot. Failure to do so will make the system unstable!

SFP’s.

System Firmware Version 5.1.1.7 Page 59

Release 4.2.2.3

 No Updates

Release 4.2.1.3

 Please refer Dell PowerConnect Users Guide for more details

Release 4.2.0.4

 Please refer Dell PowerConnect Users Guide for more details

Release 4.1.1.9

 No Updates

Release 4.1.0.19

The default authentication profile has chang ed from "enableList" to "enableNetList:

Telnet and SSH default to using the enab le NetList authentication pr ofile which requires an enable or line password. The serial console default s to using the enableList authentication pr ofile which does not require the use of a password. This change increases compatibility with industry standard behaviors. In previous releases, telnet, SSH and the serial console defaulted to using the enableList profile which does not require a password.

The following table lists changes to the User’s Configuration Guides after publication:

Change Affected Platform

Added examples of how a user would implement VLAN or QoS policy assignment in the Port Based Security section.

Added a table of the RADIUS attributes supported by the PowerConnect switches in the Controlling Management Access section.

The valid VLAN range is described as 1 to 4093 (multiple sections) PC70xx A better description of the RADIUS 'deadtime' parameter is added to the Controlling

Management Access sectio n The CLI command "monitor session 1 mode" is described in the Configuring Port Mirroring

section

All Platforms

60 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

PC8100 - SCP or SFTP

SCP or SFTP file transfer from PowerConnect

iSCSI connections to drop.

8024 - Routing issue,

Some null mac address (0000.0000.0000) present

PCM6220 v5.0.1.3

monitored

PCM6220 Switch crashes every 30 to 40 days

M6348 Stack - interface

physical interface is up

Due to a race condition, while the lag member

existing members in the lag goes down.

Summary

User Impact

Workaround

Show A AA IAS-USERS

External CDP/ISDP

Occasionally external CDP/ISDP packets are

connected neighbors appear to be seen.

None.

Multicast sources that

If an intermittent multicast source that has been

receiving the multicast stream.

Known Issues

The following are all the outstanding known issues from previous releases. The issues listed here may have been discovered on any of the switches listed on the title page.

Release 5.1.1.7

Summary User Impact Workaround

file transfers to FreeBS D and Mac causes switch to crash and reboot.

8100 switches to a FreeBSD or Mac client/server causes the switch to crash and reboot. Eventually connectivity comes back after a few seconds, but the connectivity loss is long enough to cause

Workaround: None.

Fix is under review and it will be available in the next maintenance release.

null address in ARP table after VLAN manipulation

switch crashes every 3040days while SNMP

in static LAG goes into inactive state while

Release 5.1.0.1

Molex QSPF DAC Cable with part number 111040-1104 does not comply with QSFP specification SFF-8436. These cables do not support 'vo ltage' diagnostics.

<Username> Command Missing

in PC8024 ARP table. This happens after vlan manipulation (like creat e, delete, define IP address, add and remove on trunk port).Issue occurs randomly and not immediately (take between 30min and 1h between vlan manipulation and routing issue)

when SNMP monitoring with cacti version 0.8.8a

links flap, the lag member that linked up may be put into standby, when at the same moment any

Voltage is displayed as 0.00 instead of "N/A" for this diagnostic par ameter.

The “show aaa ias-users [username]” command seems to have been deprecat ed even though it still exists in the CLI guide.

Workaround: Reloading the switch resolves the issue temporally (till next vlan reconfiguration). Add static entry in ARP table resolves the issue.

Fix is under review and it will be available in the next maintenance release.

Workaround: None. Fix is under review and it will be available in

the next maintenance release. Workaround: None.

Fix is under review and it will be available in the next maintenance release.

Ignore the voltage displayed field for this part or use a SFF-8436 compliant cable.

The same information can be seen within the running configuration of the switch with the “show running-config” command.

traffic occasionall y forwarded onto internal ports

cease sending multicast are timed out and removed from the multicast forwarding cache after 150 seconds

System Firmware Version 5.1.1.7 Page 61

being forwarded to the internal ports. This results in confusing information from the blade server point of view as multiple directly

aged out of the multicast forwarding cache begins sending again before the corresponding S,G entry has timed out at the RP (185 seconds per RFC 4601), any *,G entries (joined hosts) may take up to one IGMP Query interval to begin

The default IGMP query interval is 125 seconds. In practice, this situation is very unlikely to occur as a multicast source that fails to send even one packet for 150 seconds is unlikely to start sending packets before the S,G entry at the RP times out.

Release 5.0.0.4

Summary

User Impact

Workaround

QSFP Base CR4 with

When using the DAC copper cables with QSFP

On PC8100 switches,

to the switch.

This is a corner case issue r equiring many steps

Failure log messages

on root node.

There is no effect besides t he annoyance of these

Speed is not applied to

Since these configuration variables exist for both

speed or duplex.

SFP+ module is still

QSFP to 4x10G

The information that will be stored in the Status is not shown after

HTTPS session

The user may not be aware that the download has

Console logs/syslogs are

preference.

This will only be seen when disabling the local

The following unwanted

logs

There is no effect to the switch or user, just

The following unwanted

alloc” logs

There is no effect to the switch or user, just

copper DAC cable lengths greater than 3 meters will not work on PC8100 switches

sharp decrease in 6to 4 tunnel traffic is observed when ipv4 static route is deleted and added back

may show up on console on doing a SNMP walk

combo (copper) ports after saving and reloading the DUT (8024/8024F).

detected in runningconfig after clear config on PC8100 with

CR4s (which are rarely used) and cable more than 3 meters, the switch may not even detect this device and so link up never happens. Only the following cables are recommended with thes e CR4 QSFPs:

111040-1104 – 1m Passive Copper Assembly 111040-1204 – 3m Passive Copper Assembly

to get to this state but if the state is reached, the user will see less through-put on t he tunne l .

messages while doing an SNMP walk.

copper and fiber modes (combo ports), the speed and duplex are not saved since the port assumes that it will only save fiber parameters and thus the customer will notice that the port autonegotiates after reload instead of goes to a stati c

running-config specific to the 4x10G port (QSFP to 4x10G) will not be cleared after the clear config command.

None

Stop traffic for a minute or two so that the stale entry in the Linux stack is cl ear ed.

None

Always reconfigure the speed on copper combo ports after reboots on the 8024 / 8 024F.

User can negate these co mmands manually after a clear config.

the image upload using HTTP through IPv4

not generated if we disable LAG local

logs may come up on console after running the “clear config” command.

1) dot1s helper

logs

2) snooping logs

3) “ATP RX:

Failed to alloc”

logs may come up on console after running the “show tech-support” command with UDLD and MSTP configured: “ATP RX: Failed to

62 System Firmware Version 5.1.1.7

completed when using the GUI.

LAG config and will not affect the system.

unwanted log messages on console.

User must refresh screen or go to version screen for indication that the download has completed.

None

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Workaround

On PC8100 switches,

tunnel source.

This only happens after r epeated attempts at

'out of mBlks' logs are

line rate

There is no effect to the switch or user, just

Invalid characters are

statistics” command.

This only happens with this CLI command output

LPI samples are not

GUI.

This can be confusing to the user since the switch

"show fiber port op ti c a l-

This field does not apply to QSFP transceivers

Image uploaded with

networks are very slow.

This might require the User to have to retry the

Use alternate transfer met hods like TFTP, FTP,

Console messages shows

WARNING" often.

Our customers requested this warning to indicate

sensors.

Configure the logging level to be higher than

Finisar LRM 10G SFP+

packets intermittently.

This specific model is not recommended for use

Please use the Avago 10 G-BaseLRM model – OpenManage displ ays

duration

OpenManage timeout warning window is

None.

Command "show fiber-

transceiver count.

Executing the command “show fiber-ports

None. Port goes down if the

feature enabled on it.

If the port is operating in Trunk mode and UDLD

Include the port in na ti v e v la n. SNMP V3 walk may fail

command is issued.

After failover, SNMP V3 walk may fail, as MIB

stop intercommunication.

After a failover, close th e browser and re-open

console gets locked for 20-30 seconds if trying to add a 6to4 tunnel source within 30 seconds of deleting another

observed on console if ICMPv6 echo reques t s are sent to OOB port at

shown under reason column for some entries in the “show ip ospf

shown for the first time on selecting a different unit number using the

transceiver" command i s showing current as zero for QSFP transceivers.

adding and deleting a tunnel source and thus the impact is very small to th e customer.

unwanted log messages on console.

and thus might be bothersome.

will not show the stats for the port after a new unit is selected but then the stats will show up after going to new port and then back.

and should display “NA” instead of zero.

Wait for 30 seconds after deleting the tunnel source and then add a new tunnel source.

None.

The Web interface will not show these characters.

Select a random port from the web and then reselect the desired port to view the LPI samples.

None.

HTTP method can't be downloaded back to the switch if switch and PC are located in different networks and the

"Thermal state raised to

transceiver model – FTLX137D3BCL - drops

incorrect session timeout

ports optical-transceiver" causes UI to become slower with multiple

port is excluded from native vlan with UDLD

after “initiate failover”

download or change the IP address or VLANs on the switch or PC.

better visibility into the Thermal settings and

with our switches.

displaying the wrong value but uses the set value.

optical-transceiver” with multiple transceivers inserted will cause the console prompt a long time to return (possibly up to 60 seconds).

is enabled, the port goes down if it is excluded from the native vlan.

browser and agent can become out of time and

etc.

“Warning”.

AFBR-707SDZ-D1.

it. SNMPv2 works correctly.

System Firmware Version 5.1.1.7 Page 63

Summary

User Impact

Workaround

OOB static IP entry

address

After receiving an IP address from the DHCP

gateway that is the same as the DHCP IP address.

Use a different IP addres s than the DHCP

Summary

User Impact

Workaround

Stack member units port web GUI.

Wrong id for the power LED is displayed.

WebUI sFlow Polling

sFlow>Sampler configuration and sFlow>Poll 1/0/9.

"show interfaces

value

Some general mode VLANs tagged to the

M6220 OOB interface

failover

On a failover on a M6220 stack the OOB

CLI command "show ip

information.

"show ip dhcp snooping binding" not displaying

counts.

OOB interface

though told not to do so.

The setup wizard does not honor selection to not

DHCPv6 - M6220,

Relay

By looking at the counters, the client believes it File modification date &

time & date.

When internal files are modified, they do not get SNMP walk is not

walk on root port.

SNMP walk on the root port times out.

FIP sessions are getti ng

Users are unlikely to ch ange either the default

cannot be the same as the DHCP assigned IP

server, the switch gives an er r or and will not allow an attempt to add a s tatic IP address and

address given.

Release 4.2.2.3

display is missing in

page issues.

Configuration don't work for ports Gig1/0/1 -

None

Release 4.2.1.3

Summary User Impact Workaround

switchport " command shows incorrect General Mode Tagged VLANs

unreachable after

dhcp snooping binding" not displaying client

interface may not be displayed using this command.

interface may not be avail able.

client information. "show ip dhcp snooping statistics" does not display correct message

Use “show vlan” command to interfaces. No other workaround.

Recovers when the master is restored.

None

configured with default IP via Setup Wizard

PC/M 8024/ k - Client Solicits do Not Seem to Make it to the Server /

Release 4.2.0.4

Summary User Impact Workaround

time is not getting updated with current

successful while doing

disconnected for VLAN switchport mode change with default vlan settings.

setup OOB IP address.

is sending solicits; but, the relay and the server never see the solicitations.

the current time stamp.

VLAN or the FIP VLAN while sessions are active.

OOB IP address will need to be setup manually.

None

Do not do SNMP walks using root port.

Do not change the FIP VLAN while sessions are established. The reason is that after a VLAN is re-assigned or even changed to general mode, the sessi ons will of course be disrupted since the FIP snooping occurs on the VLAN and any change in this will cause problems on the Cisco Nexus side and will require an interruption and reestablishment of the addressing from the CNAs on up.

64 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Workaround

OpenManage displ ays

After the browser session timeout (default 10

Cannot disable the SFS

Even with SFS allow-downgrade disabled, the

Console port locks up

Radius authenticated Telnet login attempt when

expires.

Flowcontrol is in inactive

combo ports.

Flowcontrol gets automatically disabled and if

7048R replacement

Although the power supplies part number is the

supplies still works correctly.

PowerConnect 7048P :

control

After a long period of time and numerous IP

filling the logs.

SNMP showing packets

doesn't show the same.

SNMP monitoring tool is pulling stats from

but switch doesn't say the same.

PC M6220 Running

master, 4.1.0.19, 4.1.1.7

When ma s t e r member wa s removed from and

Stacked m8024-k or

Customers downgrading a switch stack to a

Failure to follow these i nstructions may cause

time.

The stack will stop passing traffic if all switches

Jumbo frames cause

Transmit Errors is seen on port channel, but not

incorrect session timeout duration when usi ng secure HTTP

"allow-downgrade" feature

when awaiting telnet Radius response

state when connected to Partner(Randall) with

power supplies reported as incompatible

poe_lldp.c(1741) 23137 %% Failed to get pairs

minutes), OpenManage displays incorrect session timeout durati on in the message "Your session had been inactive for more than 240 minutes", which is the actual setting for telnet and ssh. However, the actual ti meout occurs after the default period of 10 minutes, so the GUI message is incorrect.

master will still push the older firmware to a stack member running a newer version of code.

Radius daemon stopped, causes console port to be temporarily blocked until Radius timeout

the ports have a need for Flowcontrol, there will be no pause frames sent.

same, the 7048R reports it as incompatible and thus the error message is incorrect. The power

phones connected and with at least 5 switches connected as a stack, PoE error messages start

Disregard the time in this GUI message.

Always make sure the master h as the wanted version loaded.

Wait for timeout or ensure Radius server is reachable.

Re-enable Flowcontrol on Randall switch.

Ignore the error message, since there are no functional issues with this power supply.

Maximum supported IP phones at this time is

64.

discarded while cli

configuration altered after removal of stacking

8024/8024F switches running code that supports Ethernet stacking should not be downgraded to ol de r code that doesn’t support Ethernet stacking.

PC7048P, in a stack will error out when all rebooted at the same

transmit errors on port channel

M8024 and shows very high discard receive rate

reinserted into an M1000e enclosure, some of the configuration was missing from running-config.

version of code that does not support stacking are advised to:

1) Break apart the stack into individual units

2) Clearing the saved config on each unit

3) Renumber each unit to unit 1 using the switch X renumber 1 command

in the stack are rebooted simultaneously.

on the members of the port channel. Counters remain inaccurate.

None.

Always write the config before removing master.

units to crash when booted into the 4.1 firmware. To recover the switch, attach a serial cable and enter the boot menu (press 2 - Start Boot Menu at the prompt). From the boot menu, select 10 - Restore configuration to factory defaults, and then select 1 - Start operational code

If this situation happens, it is recommended that each power switch be r ebooted in sequence with a four to five minute interval between each reboot.

None – Inaccurate transmit errors do not cause functional issues.

System Firmware Version 5.1.1.7 Page 65

Summary

User Impact

Workaround

PC8024F St a c king -

CLI/WebUI shows the stack-ports counters value

CLI command "no snmp-

group" doesn't work

CLI command “no snmp-server community-

Dot1x ias local

The internal datab as e feature, which is a Dell

time.

"ipv6 pim join-prune

working

CLI command “"ipv6 pim join-prune interval 30"

Incorrect status being

When PowerConnect 8024F com bo por ts 21 and

Down.

Power supply

There is no impact to the user.

M8024-k Internal port s

switch POST

Network operators may exp er ience a single

None

M8024, M6220,

Submittal Troubles

CSR may be rejected by Certificate Authority

Users can hand edit the CSR to left align all

Sflow conf ig ur a ti ons do

The operator entered a s flow command that takes

time period may have already expired.

The configuration will show in the running Password strength -

Operator may become confused regarding switch

As is documented in the CLI Reference

strength minimum character-classes

CLI/WebUI : stack-ports counters value is alwa ys zero

server community-

authentication method not working

interval" command is n ot

reported with show interfaces commands

descriptions should be modified as "Internal" and "Remote" or "External" for main and secondary power supplies respectively in show system.

as zero after stack is formed.

group” is not available.

specific feature, is not working correctly because this feature requires a user to authenticate using MD5, which is not supported by Windows at this

does not change the default interval value from

22 are used for stacking, the show interfaces status command and sho w interfaces media-type commands report the status to be Detac he d and

None – Stack-Ports counter values as zero doesn’t affect the PC 8024F stacking functionality.

None

Use any other form of dot1x authentication.

None

are up during most of the

M8024k, PC8024F Extra Lines and Tabs in CSR Cause CA

not get saved to the config

minimum characters not enforced

66 System Firmware Version 5.1.1.7

bounce during reboots of the M8024-k.

a timeout value which makes th e sflow configuration valid only for a fixed period of time. As the configuration is temporary (bound by the timeout value), it is not shown in runningconfig. Storing the vale in the running config/startup config would be an error as the

operations.

rows and eliminate blank lines before submission to the CA.

config if the command is used with the "notimeout" option, e.g. sflow 1 destination ow ne r 1 not imeout

Manual, the password strength minimums do not become effective until the user has configured a non-zero value for passwords strength minimum character-classes. Please refer to the documentation for passwords

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Workaround

M8024-k OpenManage

Operators using IE 8 may experience issues as

(6) There are two "Slot 0" in display – confusing

Operators should consider using a browser

Inline help for

loopguard" is incorrect.

The inline help indicates that the command

Operators should b e aware that this global

show interfaces

10GBASE-T on 8024F

Operator should be aware that the type is

Ignore the type in the display for this

VLAN name config lost

When upgrading from release 3.1.4.16 to 4.2.0.4

affected.

The user can cut the VLAN na mes and save

Web GUI sFlow

Fails with an error - "Error! Failed to Set

Always configure the receiver owner when

Websense: Order of

The order that the commands “switchport mode

running-config” output.

None.

The “show dot1x

The User Guide states th at there is a “show dot1x

supposed correct command.

Use the “show dot1x interface statistics”

Summary

User Impact

Workaround

LLPF is not supported

The hardware functionality needed to support

Upgrade to a PowerConnect switch that

Web UI stack view display

"spanning-tree

advanced firmware displays ports 1-20 as

when upgrade from

3.1.5.16 to 4.2.x .x

configuration , receiver owner inconsistent with CLI

follows: (1) Status LED does not glow to identify Master (2) Member units do not show "DELL PowerConnect M8024-k" model info (3) Member units port display is missing (4) LED label should be "Unit No." instead of "Stack No." ? (5) All Stack No. shown are as "01", instead of "01", "02", "03", "04" e tc.

applies to a single port.

reported incorrectly on this platform.

the VLAN names only will not be copied to the upgraded configuration. VLAN numbers are not

'Receiver Owner' with '' However, it is possible to save the same configuration (without receiver owner) via the CLI.

Also, it is also not possible to add any polling device or sampler configuration on the Web GUI with this receiverIndex (seen in the show all option) since an error is thrown stating "Receiver Index is not configured".

other than IE 8.

CONFIG command app lies to all ports.

command.

these off before the upgrade occurs and then again reenter them after the upgrade.

using the Web interface.

switchport commands in running-config changed, Impacts scripts

statistics” command within the User Guide is actually “show dot1x interface statistics” command from the CLI.

general” and “switchport general allowed vlan” are displayed in the running config has been changed so that now the “allowed” command is first. This may cause problems with customer’s scripts that expect a certain order in the “show

statistics” command for displaying dot1x statistics on the switch when the actual command for doing this is “show dot1x interface statistics”. This can be frustrating to the customer if he has referenced the User Guide to look up the

command.

Release 4.1.1.9

on the PCM6220 switch

System Firmware Version 5.1.1.7 Page 67

LLPF is not available on the PCM6220 switch.

supports LLPF like a PCM6348 switch.

Summary

User Impact

Workaround

Traffic led solid green

Setting advertisement from other switches or IXIA

None – Traffic LED Solid Green do not cause

Config Lost for the

On boot up, the switch gives an error and rejects

Remove port configuration of 10G module

PC7XXX cable

ports does not work.

DUT delivers more

DUT may draw more power than negotiated at

None – system assumes 5.8W average loss due Trunk mode VLANs

Not compatible with other vendors trunk modes.

Administrators can configure “general” mode

Speed/duplex command s

Confusion about how to configure links.

Documentation and CLI prompt clearly states

ST : Stack member

No user impact expected. Observed occasional

None required.

vlan protocol gro up

with 1Gb and 100Mb link speeds on PC8024/PC8024F

Portchannel on removing a 2 Port 10 G Module on the PCM8024 switch

of 1Gb or 100Mb link speeds to 10G links on 8024, 8024F results in link up and traffic led to be solid green.

all the config for the portchannel as well as the TenGig interfaces i n Bay1.

any functional issues.

before removing the module from the bay and reboot the switch.

Release 4.1.0.19

Summary User Impact Workaround

Log message output snmpwalk will report error log message related to

CPU-port and vlan routing port

Secure HTTP Random Characters

Under System Management -> Secur e H TTP, random characters maybe populating some of the fields.

None – error messages do not cause functional issues.

The CLI must be used to generate certifications.

Release 4.1.0.6

Summary User Impact Workaround

diagnostics for the Fiber

power than the PD requested via LLDP in high power mode.

L3 routing NSF failover data plane on dynamic LAG - loss duration up to 5 seconds for large configurations

Fiber port cable diagn ostics are not available for the PC7XXX.

short cable lengths. PD may draw more power than negotiate d, but power loss due to cable impedance is compensated for so that devices with average or longer cable length will receive adequate power.

Interruption of voice, video and data service for duration of loss. Data plane loss during failover should not exce e d 50 ms.

None.

to cable length and del ivers 5.8W extra power to ensure device receives requested power.

Disable portfast and auto-portfast on physical ports configured in a LAG.

transmit tagged frames only

available for interfaces which require autonegotiation

response times to ICMP ping requests in a 12 unit stack are larger than for stack master

Issue with protocol based VLAN configuration migration.

68 System Firmware Version 5.1.1.7

outlier response time up to 500 ms for stack members in a large stack configuration with heavy traffic. Average response time is well under 100 ms for stack members. All response times are well within ping limits.

The command string parameter in earlier versions; now it requires an integer par amet er.

required a

VLANs, which transmit PVID frames untagged and all other VLAN frames tagged. General mode is compatible with other vendor’s trunk mode behavior.

which commands are app licable to which interfaces. Only use speed/duplex commands

on fiber interfaces. Only use speed auto/duplex aut o co m mands on copper interfaces.

The software recognizes if the group name is alphanumeric, however it will not work when the name of the group is numeric (for example 2, 3, etc.).

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Summary

User Impact

Workaround

Read/write user allowed

The user always gets Read-On ly access if using

User can configure the same TACACS user Configure the acct-port to the default using the

read only access when authentication method is used as TACACS.

TFTP gives no reason for file download failures.

CLI command stack-port config rejection does not display the cause.

The 'acct-port' command does not have 'no' version.

Non-configuration file getting loaded to startupconfig through HTTP.

A v6 ping with the v4 header destination address set to 224.0.0.2 (all routers addr) is not responded to.

Certain packets match system rules that elevate the priority for protocol packets.

TACACS as a means for HTTP authentication, even if the TACACS user is Read/Write capable.

Generic failure message i s issued. Administrators can ping the TFTP server from

If a user enters an invalid interface, a generic error message is issued .

The user can configure the acct-port to the default using the positive form of the command

Switch does not utilize invalid configuration file information. Earlier versions of startup-config are not available for fallback when overwritten with an invalid startup-config.

Users are not able to ping over 6to4 tunnels using IPv6 addresses.

Packets may be transmitted out of order when using priority flow control. Additionally, if the queue that the packets are put on is not enabled for lossless PFC, then the packets can be transmitted even when the port was told to pause. This may have an effect on connections that expect packet order to be maintained, e.g. FCoE.

locally and use LOCAL authentication method for HTTP. The user will be able to get access based on the local user access level (Read-write or Read-only).

the switch. Administrators should ensure the TFTP server is available, the requested file is available, and the permissi ons are set correctly.

Utilize the show stack-port command to identify stack port configuration issues.

acct-port 1813 command in Radius accounting mode.

In this case, an invali d configuration file was downloaded (on purpose) via the web. When the switch rebooted, it detected that the configuration file was invalid and overwrote the start-up config with the default configuration (an empty configuration). Users are advised to maintain off-line copies of switch configurations.

Users can send pure IPv4 pings to the other end of the tunnel.

None.

System Firmware Version 5.1.1.7 Page 69

Known Restrictions and Limitations

Description

User Impact

PowerConnect 8100 Series QSFP DAC CR4 40GB cable fails to link up with

S4810(conf-if-fo-0/60)#intf-type cr4 autoneg

Description

User Impact

SNMP walk is not successful while doing walk on root port

Low

Release 5.1.1.7

partner switch (for example. Dell Force10 S4810). After upgrading a Power Connect 8100 series switch to firmware 5.1.0.1 customers may experience loss of connectivity when using 40G QSFP CR4 DAC cables with partner switch. Some partner switch vendors may by default leave auto-negotiation disabled for CR4 connections. This will cause a negotiation mismatch and the QSFP CR4 port will not link up.

When https enabled with a 2048 bit key CPU sslt tasks will require a high CPU processing load for 5 – 10 seconds.

PCM6220 switch does not support “Private VLAN” feature Private VLAN feature requires hardware

High In 5.1.0.1 firmware a behavioral chan ge, as

per IEEE spec, was made th at means that when a QSFP CR4 cable is connected the 40gb port has auto-negotiation enabled. In previous versions of firmware the default behavior was to leave auto-negotiation disabled. This change was required in order to ensure compliancy to the following specifications: IEEE 802.3-2012 Secti on 6 / Clause 85.3 / Clause 82.6 / Clause 73.9 - explains that auto-negotiation is mandatory for 40GBASECR4 and 100GBASE-CR10.

Workaround: Enable auto-negotiation on 40G interface of

Dell Force10 S4810 switch like below:

S4810#conf S4810(conf)#int fortyGigE 0/60

Medium HTTPS login attempt may timeout.

Workaround: Use 1024 or lower key.

support that the PCM6220 XGS3 swi tc hi ng fabric does not contain.

Workaround: None.

Release 5.1.0.1

The IPMC L2_BITMAP is only updated to reflect ports that should 'not' be flooded when a multicast stream is 'first' seen on a VLAN.

70 System Firmware Version 5.1.1.7

SNMP walk on root port may result in a timeout if executed without specifying any timeout value.

Workaround: Execute walk with a recommen ded timeout

value of 2 sec Moderate

There will be some flooding of multicast traffic

Workaround: Issue no ip multicast followed by ip multicast

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Description

User Impact

The device manager GUI does not

Moderate – Multiple errors are displayed when trying to bring up the device GUI

Workaround: The User must use an Internet Explorer version prior to IE9 with the

"ip http secure-server" may not be cleared after performin g t he clear config operation if there is user intervention before prompt returns

Reboot needed when critical message “Failed to add local route for network X/X on interface X” appears on the console

Release 5.0.1.3

System – 5.0.1.3

Low Clear config operatio n clears all system

components operational and configuration information and then builds the default configurati on. T his ope r a tion takes time to complete and the prompt may not be returned immediately. Prompt is returned after the completion of the operation. Please wait for the prompt, pressing "enter"/CR is not needed in this case.

Workaround: Issue clear config one more time if “ip http secure-server” configu ratio n is not cleared

High Route is not added and hence critical

situation. Need a rebo ot to recover.

Stacking

Description User Impact

Under certain conditions 5 or more ports identified as stacking ports can cause transmission errors even if the stacking links are not up.

Low This problem happens infrequently and setting those ports back to Ethernet mode

has corrected the prob lem.

Release 5.0.0.4

System – 5.0.0.4

System

Description User Impact

8100 switches increment “Internal MAC Rx Errors” counter, when packets are received with (size > 1518) and (size <= MTU size), and the packet contains an invalid FCS or code error detected or an IEEE length check error

Web

support IE9 at this time.

Low For typical IP networks this problem won't be seen because t he length

field/Ethertype will contain a valid value.

with IE9.

GUI. Note: This is not a limitation if running Release 5.1.0.1 or later versions of

firmware.

System Firmware Version 5.1.1.7 Page 71

iSCSI

“show iscsi sessions” command does not

platforms.

Moderate –

traffic.

Description

User Impact

Show policy-map interface command

Low – the administrator can show the counters on the individual members of the

Show dot1x statistics command is

Low – the administrator can show the statistics on the individual interfaces

Description User Impact

display established sessions on M8024/M8024-k/PC8024/PC8024F

Release 4.2.1.3

Management – 4.2.1.3

CLI

CLI command ‘show dot1x users’ is missing in CLI

does not take a port-channel parameter

missing in the CLI Show crypto key pubkey-chain ssh -

"username" optio n doe s not work

Workaround: Configure partner devices to send tagged, not priority tagged

Low – Use the show dot1x clients command to show the authenticated clients.

port-channel and sum them manually.

None – the administrator must configure and associate a key to a user in order for the user to be associat ed with an SSH key. The following exampl e creates and associates an SSH user with a key:

console(config)#us er name asd password XXXXXXXX privilege 15 console(config)#crypto key pubkey-chain ssh console(config-pubkey-chain)#user-key asd dsa console(config-pubkey-key)#key-string "ssh-dss

AAAAB3NzaC1kc3MAAACBAJRwUAD3AuRACp1MObBeh1AgyZb18wf9Bt dip+t+1CbAqiqNEh4lBiew184DSKk0T6SnSSXuCN+bJnQPxJeiQt+OFnmjiYhn HcvI04Q5KnQhloZcEFgSsmQ7zJnReWtLvUQI0QvBIStanzedmQVGHvDrQ5X 2R729ToSH0ibBrnYtAAAAFQDNord7S9EJvUkKKxVBpWE6/skCmQAAAIB MjMO+BPP5KXzNWfZhqAhxBSoBvif/z6pzi9xWLlYy99A03zmRYCpcGIoLW iRHsR7NVpxFqwbqvez8KS0CDJ5aoKKLrpBlpg5ETkYEew/uTZ14lQQRBrzPw GBfxvTXKCWiI2j5KFa/WKLSnmWJX0/98qpxW/lMXoXsA9iK4pnMKwAAAI B4Jrt6jmoLybpzgOPOI0DsJ7jQwWacinD0jliz8k+qzCpanhd2wH+DEdj/xO2sFR fnYlME3hmXoB+7NByVUtheVjuQ2CWhcGFIKm9tbuPC6DtXh1xxT0NJ7rspv Lgb0s6y/0tk+94ZP5RCoAtLZ7wirShy3/KJ4RE0y2SFZjIVjQ=="

console(config-pubkey-key)#exit console(config-pubkey-chain)#exit console(config)#exit console#show crypto key pubkey-chain ssh username asd

Username : asd ssh-dss

Fingerprint : d9: d1:21:ad:26: 41:ba:43:b1:dc : 5c :6c :b9:57:07:6c

72 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

SSH RSA or DSA keys can be generated by using the ssh-keygen command on a

Broadcom NIC link always stays up

Low- This is works as designed and is necessary for new features for 12G.

Description

User Impact

Windows Vista® Authentication - The

Low

The maximum number of 802.1x clients

Low – most deployments will have at most 2 802.1x clients per physical port. The maximum number of configurable

Low as most deployments use a single traffic class. Some limited deployments

Description

User Impact

Maximum number of unicas t static

The maximu m number of unicast MAC and source port filtering entries is 20.

Maximum number of multicast static

The maximum number of multicast MAC and source port filtering entries is 20.

mac addr-table multicast filtering

Description

User Impact

No command to identify external IGMP

There is no specific command to identify an external IGMP querier.

Unix system or with other publicly available utilities.

within the m8024-k status even after disabling server port through device manager or turning off server altogether.

Release 4.2.0.4

Layer 2 – 4.2.0.4

802.1x Authentication

Windows Vista® client could fail to authenticate properly when the option to cache user credentials is selected.

per port is 4.

traffic classes is 7.

MAC Filtering

The Broadcom NIC 57810S never reports to the internal switch that the link is down because it needs the link to be up so that internal communications can continue with iDRAC and for other various components even after disabling the link on the server side.

Workaround:

1. In Control Panel  Network Connections, right-click on the desired

Local Area Connection and select Properties.

2. In the Properties window, select the Authentication tab.

3. Deselect the checkbo x for Cache user information for subsequent

connections to this network.

4. Click OK.

Note: If running Release 5.1.0.1 or later version of firmware, the maximu m number is 24.

may use up to 3 traffic classes.

filtering entries

Static multicast MAC address table entries do not show with show command

The maximum number of multicast MAC and destination port filtering entries is

256. Users must enable MAC filtering using the

command to enable filtering. Static MAC multicast forwarding entries will then be shown.

LACP

Description User Impact

LAGs Supported Number of LAGs supported:

128 total LAGs of which up to 72 may be dynamic LAGs. Up to 144 ports can be assigned to dynamic LAGs. The PCM8024 supports 12 LAGs with up to 24 ports assigned to dynamic LAGs.

IGMP Snooping

querier

System Firmware Version 5.1.1.7 Page 73

Administrators can use the show ip igmp snooping querier det a il command or the show ip igmp snooping querier vlan command to display information about snooping querie rs .

IP VLAN MTU Support

Operators may see jumbo packets discarded when oper ating in a routed IP

IPv6 Fragmentation Support

The switch is not fragmenting the datagram and forwards even when the IP MTU

Description

User Impact

Description

User Impact

radius-server mode commands do not

Low - None of the commands in radius-server mode support a "no" form except

Multicast VLAN Registration

Description User Impact

MVR is not supported on LAGs Use of MVR is restricted to physical interfaces.

Layer 3 – 4.2.0.4

IP MTU

Description User Impact

environment. Administrators are advised when operating in a L3 routing configuration with jumbo frames to adjus t both the link MTU and the VLAN IP MTU.

IPv6 MTU

Description User Impact

of the forwarding interface is set to a lower value (than the datagram size).

IPv6 frames are not allowed to be fragmented. IPv6 frames forwarded in silicon can be up to the lesser of 9216 octets or the link MTU. These frames are forwarded by the switching silicon with no effect. If a frame exceeds the link MTU for a port, it is discarded silently.

If a packet is sent to the CPU or originated on the CPU and it exceeds the IPv6 MTU, then the packet still will not be fragmented. Instead, an ICMP error message is returned t o t he sender. The maximum IPv6 MTU is 1500 bytes.

Administrators are advised that when operating in an L2 switching configuration with jumbo frames, to only adjust the link MTU and let the system automatically adjust the IPv4/ IPv6 MTU based on the link MTU.

VRRP

The maximu m number of VRRP instances is 50.

Users can scale VRRP higher than previously.

Management – 4.2.0.4

CLI

have a "no" form.

The maximum command line length is 1536 characters.

USB

Description User Impact

Dir command can only address top-level directory on USB stick

Only FAT32 formatted devices are supported.

for the msgauth command. To reset values to the default, delete the server entry and add it back.

Low - Entries greater than the maximum line length throw an error, e.g. using multiple interface range qualifiers.

Minimal – users can move files to top-level directory easily

Minimal – FAT32 devices are the de-facto standard for flash devices

74 System Firmware Version 5.1.1.7

PowerConnect 8132/8164/8132F/8164F/8024/8024F/M8024/M8024-k/

M6220/M6348/7024/7048/7024P/7048P/7024F/7048R/7048R-RA

Release Notes

Description

User Impact

Certain browser (IE) versions respond

This behavior is a browser performance limitation . Users may select anot her Certain browser (FireFox) versions

This behavior is a browser functionality issue. If popups are blocked, the web

Description

User Impact

CLI Comment Character

The '!' indicates the beginning of a comment. All characters following the '!' will

Description

User Impact

VLAN request/Notification counters are

Low – FIP ses s ions are established and are d isplayed in the show commands.

equest/Notification counters are not incremented for priority

Broadcom CNA only supports

Low – Most users prefer automatic setup of CNAs

be able to do so if using BRCM CNAs.

While working with QLogic, it is found that CNA sends two TLVs concerning the

be a use case where host or FCF can choose to create di fferent priorities for

When multiple partitions are present on the flash drive, only the first partition is accessible.

Web

Minimal – users will typically re-partition flash drives to maximize space.

slowly when displaying large lists of information. In these cases , the “All” display selection may not appear (is disabled).

automatically block popups after a certain number of displays within a session.

supported brow ser to enable “all” display functionality. Alternatively, the user may utilize the page selector functions to display the appropriate page of information.

interface will display errors/information using alerts. Users can disable popup monitoring by browsing to about:config and set dom.popup_maximum to -1

File Management

be treated as a comment (except when configuring a banner where the ! is accepted at the beginning of a line)

Data Center – 4.2.0.4

Interoperability

not incremented for priority tagged packets

configuration of VLAN via VLAN Discovery process

FIP snooping VLAN request or VLAN response counters do not increment when CNA sends priority tagged frames (i.e. VID 0). It is observed that certain CNAs sometimes sends FIP VLAN discovery frames as 802.1Q tagged frames with VID=0 and priority as FCoE priority. FSB snoops FIP packets which are classified to any of FCoE enabled VLANs. In this case, the default VLAN (1) and the configured FCoE VLAN. In general, VLAN request messages are exchanged as untagged so gets associated with VLAN 1. In this case, the BROADCOM CNA sends priority tagged FIP frames (VID=0) which do not match the FSB snooping criteria because they do not get associated to any of the F C oE enabled VLANs hence FCoE VLAN R tagged packets. These con trol frames are however are forwarded to FCF by FSB in order to establish a successful session. FIP snooping bridge need to consider forwarding priority tagged FIP frames to FCF as well in order to interoperate with CNAs which could send priority tagged packets. The current FSB implementation does this already.

The Broadcom CNA does not support manual configuration of VLAN. This means that customers who prefer to set up their FCoE network manually will not

FIP snooping bridge does not forward the DCBX FIP tlv info from configuration source

System Firmware Version 5.1.1.7 Page 75

Low – No supported FCF supports this configuration option.

FIP/FCoE traffic. It sends FIP TLV along with FCoE TLV with respective priorities - in this case same priority for both. Although, it is not a normal use case but can be used to have different treatment for control and data traffic. Ther e can

control (FIP) and FCoE data traffic.

FIP Snooping Over Stack Results in

Low - When trying to accomplish FIP snooping over the stack, the eNode address

supported across the stacks so this is a non-issue.

In cases, where FIP and FCoE use different priorities, it is expected that intermediate switches are configured to treat them accordingly. In cases where ports are configured in DCBX auto mode and configuration source carries two TLVs, one for FIP and other for FCoE with different priorities then it is expected that FSB/DCB should forward this information to downstream ports.

The PowerConnec t FS B implementation ignores the FIP TLV and does not forward this information to the peers. In such cases, Host will not know the special treatment or expected priority for FIP frames. Considering that slow protocol and this being a corner case situation, this can be a readme and documented in release notes.

eNode in Ethernet VLAN

is put in the wrong VLAN causing the connections to never occur. FCoE is not

End of Release Notes

76 System Firmware Version 5.1.1.7

Dell PowerConnect 7024P, PowerConnect M6348, PowerConnect 7048R, PowerConnect 7024F, PowerConnect 8024F User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Table of Contents

Introduction