System Management Configuration Guide for the Cisco NCS 6000
Series Router, Release 5.0.x
First Published: 2013-09-01
Last Modified: 2014-01-01
Americas Headquarters
Cisco Systems, Inc.
170 West Tasman Drive
San Jose, CA 95134-1706
USA
http://www.cisco.com
Tel: 408 526-4000
800 553-NETS (6387)
Fax: 408 527-0883
THE SPECIFICATIONS AND INFORMATION REGARDING THE PRODUCTS IN THIS MANUAL ARE SUBJECT TO CHANGE WITHOUT NOTICE. ALL STATEMENTS,
INFORMATION, AND RECOMMENDATIONS IN THIS MANUAL ARE BELIEVED TO BE ACCURATE BUT ARE PRESENTED WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED. USERS MUST TAKE FULL RESPONSIBILITY FOR THEIR APPLICATION OF ANY PRODUCTS.
THE SOFTWARE LICENSE AND LIMITED WARRANTY FOR THE ACCOMPANYING PRODUCT ARE SET FORTH IN THE INFORMATION PACKET THAT SHIPPED WITH
THE PRODUCT AND ARE INCORPORATED HEREIN BY THIS REFERENCE. IF YOU ARE UNABLE TO LOCATE THE SOFTWARE LICENSE OR LIMITED WARRANTY,
CONTACT YOUR CISCO REPRESENTATIVE FOR A COPY.
NOTWITHSTANDING ANY OTHER WARRANTY HEREIN, ALL DOCUMENT FILES AND SOFTWARE OF THESE SUPPLIERS ARE PROVIDED “AS IS" WITH ALL FAULTS.
CISCO AND THE ABOVE-NAMED SUPPLIERS DISCLAIM ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, THOSE OF
MERCHANTABILITY, FITNESS FOR A PARTICULARPURPOSE AND NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE, OR TRADE PRACTICE.
IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING, WITHOUT
LIMITATION, LOST PROFITS OR LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS MANUAL, EVEN IF CISCO OR ITS SUPPLIERS
HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
Any Internet Protocol (IP) addresses and phone numbers used in this document are not intended to be actual addresses and phone numbers. Any examples, command display output, network
topology diagrams, and other figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses or phone numbers in illustrative content is unintentional
and coincidental.
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: http://
www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership
relationship between Cisco and any other company. (1110R)
Obtaining Documentation and Submitting a Service Request ix
Upgrading FPD 1
FPD 1
Prerequisites for FPD Image Upgrades 2
Overview of FPD Image Upgrade Support 2
Automatic FPD Upgrade 2
FPD upgrade service 2
Determining Upgrade Requirement 3
Automatic FPD upgrade 3
Manual FPD upgrade 3
FPD upgrade 4
Additional References 5
CHAPTER 2
Process Placement 7
Prerequisites for Configuring Cisco IOS XR Process Placement 7
Information About Cisco IOS XR Process Placement 8
What Is a Process? 8
What Is Process Placement? 8
Default Placement Policy 8
Reasons to Change the Default Process Placement 9
Reoptimizing Process Placements 9
Reconfiguring Process Placements 9
Recommended Guidelines for Process Placement 10
Process Placement Based on Memory Consumption 10
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
iii
Contents
Changing Process Affinities 10
affinity location set 10
affinity location type 10
affinity program 11
affinity self 11
Hierarchical Placement Policy 11
How to Configure Cisco IOS XR Process Placement 11
Reoptimizing Process Placement 11
Setting Memory Consumption Thresholds 12
Creating a Location Set Affinity 13
Creating a Location Type Affinity 15
Creating a Program Affinity 16
Creating a Self Affinity 18
CHAPTER 3
CHAPTER 4
Configuration Examples for Process Placement 19
Additional References 20
Configuring Manageability 23
Information About XML Manageability 23
How to Configure Manageability 24
Configuring the XML Agent 24
Configuration Examples for Manageability 25
Enabling VRF on an XML Agent: Examples 25
Additional References 25
Implementing NTP 27
Prerequisites for Implementing NTP on Cisco IOS XR Software 27
Information About Implementing NTP 28
How to Implement NTP 29
Configuring Poll-Based Associations 29
Configuring Broadcast-Based NTP Associates 31
Configuring NTP Access Groups 33
Configuring NTP Authentication 35
Disabling NTP Services on a Specific Interface 37
Configuring the Source IP Address for NTP Packets 38
Configuring the System as an Authoritative NTP Server 40
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
iv
Contents
Updating the Hardware Clock 41
Verifying the Status of the External Reference Clock 42
Examples 43
Configuration Examples for Implementing NTP 43
Additional References 46
CHAPTER 5
Implementing Physical and Virtual Terminals 49
Prerequisites for Implementing Physical and Virtual Terminals 50
Information About Implementing Physical and Virtual Terminals 50
Line Templates 50
Line Template Configuration Mode 50
Line Template Guidelines 51
Terminal Identification 51
vty Pools 51
How to Implement Physical and Virtual Terminals on Cisco IOS XR Software 52
Modifying Templates 52
Creating and Modifying vty Pools 53
Monitoring Terminals and Terminal Sessions 55
Craft Panel Interface 56
Configuration Examples for Implementing Physical and Virtual Terminals 56
Additional References 58
CHAPTER 6
Implementing SNMP 61
Prerequisites for Implementing SNMP 62
Restrictions for SNMP Use on Cisco IOS XR Software 62
Information About Implementing SNMP 62
SNMP Functional Overview 62
SNMP Manager 62
SNMP Agent 62
MIB 62
SNMP Notifications 63
SNMP Versions 64
Comparison of SNMPv1, v2c, and v3 65
Security Models and Levels for SNMPv1, v2, v3 66
SNMPv3 Benefits 67
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
v
Contents
SNMPv3 Costs 68
User-Based Security Model 68
View-Based Access Control Model 68
MIB Views 69
Access Policy 69
IP Precedence and DSCP Support for SNMP 69
How to Implement SNMP on Cisco IOS XR Software 69
Configuring SNMPv3 69
Configuring SNMP Trap Notifications 71
Setting the Contact, Location, and Serial Number of the SNMP Agent 72
Defining the Maximum SNMP Agent Packet Size 73
Changing Notification Operation Values 74
Setting IP Precedence and DSCP Values 75
CHAPTER 7
Configuring MIB Data to be Persistent 76
Configuring LinkUp and LinkDown Traps for a Subset of Interfaces 77
Configuration Examples for Implementing SNMP 79
Configuring SNMPv3: Examples 79
Configuring Trap Notifications: Example 82
Setting an IP Precedence Value for SNMP Traffic: Example 83
Setting an IP DSCP Value for SNMP Traffic: Example 84
Additional References 84
Configuring Periodic MIB Data Collection and Transfer 87
Prerequisites for Periodic MIB Data Collection and Transfer 87
Information About Periodic MIB Data Collection and Transfer 88
SNMP Objects and Instances 88
Bulk Statistics Object Lists 88
Bulk Statistics Schemas 88
Bulk Statistics Transfer Options 88
Benefits of Periodic MIB Data Collection and Transfer 89
How to Configure Periodic MIB Data Collection and Transfer 89
Configuring a Bulk Statistics Object List 89
Configuring a Bulk Statistics Schema 90
Configuring Bulk Statistics Transfer Options 92
Monitoring Periodic MIB Data Collection and Transfer 95
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
vi
Contents
Periodic MIB Data Collection and Transfer: Example 96
CHAPTER 8
Implementing CDP 97
Prerequisites for Implementing CDP 97
Information About Implementing CDP 98
How to Implement CDP on Cisco IOS XR Software 99
Enabling CDP 99
Modifying CDP Default Settings 100
Monitoring CDP 101
Examples 102
Configuration Examples for Implementing CDP 104
Additional References 104
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
vii
Contents
viii
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
Preface
This guide describes the System Management configuration details for Cisco IOS XR software. This chapter
contains details on the changes made to this document.
Changes to this Document, page ix
•
Obtaining Documentation and Submitting a Service Request, page ix
•
Changes to this Document
Table 1: For NCS 6000 Series Router
SummaryDateRevision
Initial release of this document.November 2013OL-30990-01
Obtaining Documentation and Submitting a Service Request
For information on obtaining documentation, using the Cisco Bug Search Tool (BST), submitting a service
request, and gathering additional information, see What's New in Cisco Product Documentation.
To receive new and revised Cisco technical content directly to your desktop, you can subscribe to the What's
New in Cisco Product Documentation RSS feed. RSS feeds are a free service.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
ix
Obtaining Documentation and Submitting a Service Request
Preface
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
x
Upgrading FPD
In general terms, field-programmable devices (FPDs) are hardware devices implemented on router cards
that support separate software upgrades. A field-programmable gate array (FPGA) is a type of programmable
memory device that exists on most hardware components of the router. The term FPD has been introduced
to collectively and generically describe any type of programmable hardware device on FPGAs. Cisco IOS XR
software provides the Cisco FPD upgrade feature to manage the upgrade of FPD images .
For complete descriptions of the FPD commands listed in this module, see Related Documents, on page
5. .
Table 2: Feature History for Upgrading FPD Software on Cisco IOS XR Software
ModificationRelease
This feature was introduced.Release 5.0.0
Support for parallel FPD upgrade for power modules.Release 6.4.1
CHAPTER 1
FPD
This module contains the following topics:
FPD, page 1
•
Prerequisites for FPD Image Upgrades, page 2
•
Overview of FPD Image Upgrade Support, page 2
•
FPD upgrade service, page 2
•
Additional References, page 5
•
An FPD is a field programmable logic device which contains non-volatile, re-programmable memory to define
its internal wiring and functionality. The contents of this non-volatile memory are called the FPD image or
FPD firmware. Over the lifespan of an FPD, FPD firmware images may need upgrades for bug fixes or
functionality improvements. These upgrades are performed in the field with minimum system impact.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
1
Prerequisites for FPD Image Upgrades
Prerequisites for FPD Image Upgrades
Before upgrading the FPD on your router you must install and activate the fpd.rpm package.
This is for the manual upgrade using the upgrade hw-module FPD command.
Overview of FPD Image Upgrade Support
An FPD image is used to upgrade the software on an FPD.
FPD versions must be compatible with the Cisco IOS XR software that is running on the router; if an
incompatibility exists between an FPD version and the Cisco IOS XR software, the device with the FPGA
may not operate properly until the incompatibility is resolved.
Related Topics
show hw-module fpd Command Output: Example
Upgrading FPD
Automatic FPD Upgrade
FPD auto-upgrade can be enabled and disabled. When auto FPD is enabled, it automatically updates FPDs
when a SMU or image changes, including an updated firmware revision. Use the fpd auto-upgrade command
to disable or enable auto-fpd.
FPD upgrade service
The main tasks of the FPD upgrade service are:
FPD image version checking to decide if a specific firmware image needs an upgrade or not.
•
Automatic FPD Image Upgrade (if enabled).
•
Manual FPD Image Upgrade using the upgrade hw-module fpd command.
•
Invoke the appropriate device driver with a name of the new image to load.
•
An FPD image package is used to upgrade FPD images. The install activate command is used to place the
FPD binary files into the expected location on the boot devices.
Supported Upgrade Methods
RemarksMethod
Upgrade using CLI, force upgrade supported.Manual Upgrade
Auto Upgrade
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
2
Upgrade using install SMU activation or during image upgrade. User
can enable/disable auto upgrade feature.
Upgrading FPD
Determining Upgrade Requirement
Use the show hw-module fpd command to determine if an FPD upgrade is required. Check for NEED UPGD
in the Status column.
Use the show fpd package command to find out which FPGAs are supported with your current software
release and minimum hardware requirements for each module.
Automatic FPD upgrade
Use the fpd auto-upgrade enable command to enable the auto upgrade feature.
The FPD images are upgraded as part of the install activation of the new image. The FPDs are upgraded before
the router is reloaded.
During an FPD auto-upgrade, the installed FPD rpm package includes an FPD image with a new version of
software that is different than the version of the image running on the hardware. Once the FPDs have been
upgraded, even if the base image is rolled backed to the older version, the FPD will not be downgraded to its
previous version.
When a reload package is installed with new FPD images, the FPD images are upgraded before the router
gets reloaded. This feature is controlled through an fpd auto-upgrade configuration option. The auto-upgrade
feature does not address the following:
Determining Upgrade Requirement
FPD Upgrade during initial boot
•
FPD Upgrade during new card insertion
•
Manual FPD upgrade
Manual FPD upgrade is performed using the upgrade hw-module fpd command. All cards or all of FPGA
in a card can be upgraded. If reload is required to activate FPD , the upgrade should be complete. All line-cards,
fabric cards and RP cards cannot be reloaded during the process of the FPD upgrade.
FPD upgrade is transaction-based:
Each fpd upgrade cli execution is one transaction
•
Only one transaction is allowed at any given time
•
One transaction may include one or many FPD upgrade(s)
•
The force option can be used to forcibly upgrade the FPD (regardless of whether it is required or not). It
triggers all FPDs to be upgraded or downgraded. The force option can also be used to downgrade or upgrade
the FPGAs even after the version check.
In some cases, FPDs can have primary and backup images.Note
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
3
FPD upgrade
FPD upgrade
Upgrading FPD
The key to understanding the FPD output is that nodes can have two firmware versions. One, which is currently
running, and a downloaded version, which will become the running version after the next boot. The running
version and downloaded version can be the same. There are circumstances where this is not the case, and that
would be if a node was recently upgraded, and requires a reboot to load the new updated package. Generally,
the downloaded version is the latest version, when compared to the running version. FPD packages that do
not require a reload to activate the new firmware version would not see the version skew. Below is a sample
output showing version skew on the CCC FPGA. After reload, both running and downloaded versions will
be the same. CCC Power-On is in need of an upgrade. To see what version is expected, issue the command
show fpd package and find the FPD device for that card type which is in need of upgrade.
To upgrade an fpd device, such as the one above, use the upgrade hw-module location 0/0 fpd CCC\
Power-On command or if it is more desirable to upgrade all components that need upgrading at the same
time, use the upgrade hw-module location all fpd all command. Note that this upgrade will require a reload
of the node to take effect. Adding the force option will upgrade all FPD devices regardless if they require
upgrading or not. This is not recommended.
The command used for upgrade is: upgrade hw-module location location of node fpd fpd device
The show fpd package command displays 4 very critical pieces of information with regard to firmware that
is imbedded in the current running XR image. The first column displays whether a reload would be required
to make the updated FPD version the running version. The second column shows the version number of
firmware residing on the running XR image. The forth and fifth columns show, based on the current running
XR image, what the minimum requirements are for both firmware and hardware versions for each programmable
device.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
4
Upgrading FPD
Additional References
The following sections provide references related to FPD software upgrade.
Related Documents
Cisco IOS XR command master list
Additional References
Document TitleRelated Topic
Cisco IOS XR FPD upgrade-related commands
Initial system bootup and configuration information
for a router using the Cisco IOS XR Software.
Information about user groups and task IDs
Standards
No new or modified standards are supported by this
feature, and support for existing standards has not
been modified by this feature.
MIBs
System Management Command Reference for Cisco
NCS 6000 Series Routers
Configuring AAA Services on module of System
Security Configuration Guide for Cisco NCS 6000
Series Routers
TitleStandards
—
MIBs LinkMIBs
There are no applicable MIBs for this module.
RFCs
No new or modified RFCs are supported by this
feature, and support for existing RFCs has not been
modified by this feature.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
To locate and download MIBs for selected platforms
using Cisco IOS XR Software, use the Cisco MIB
Locator found at the following URL: http://cisco.com/
public/sw-center/netmgmt/cmtk/mibs.shtml
TitleRFCs
—
5
Additional References
Upgrading FPD
Technical Assistance
LinkDescription
The Cisco Technical Support website contains
thousands of pages of searchable technical content,
including links to products, technologies, solutions,
technical tips, and tools. Registered Cisco.com users
can log in from this page to access even more content.
http://www.cisco.com/cisco/web/support/index.html
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
6
Process Placement
This module describes conceptual information and configuration tasks for process placement on your router.
Process Placement on Cisco IOS XR software balances application processes between the available based
on memory usage and other criteria. Use the procedures described in this document to reoptimize the placement
of processes, or override the default placement policies.
For complete descriptions of the process placement commands listed in this module, see Related Documents,
on page 21. .
Table 3: Feature History for Configuring Cisco IOS XR Process Placement
This module contains the following topics:
CHAPTER 2
ModificationRelease
This feature was introduced.Release 5.0.0
Prerequisites for Configuring Cisco IOS XR Process Placement , page 7
•
Information About Cisco IOS XR Process Placement, page 8
•
How to Configure Cisco IOS XR Process Placement, page 11
•
Configuration Examples for Process Placement, page 19
•
Additional References, page 20
•
Prerequisites for Configuring Cisco IOS XR Process Placement
Note
Only processes that are identified in Cisco IOS XR software as placeable can be controlled through process
placement configuration. Nonplaceable processes are not affected by placement policy. To learn the
processes that are placeable, issue the show placement program all command.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
7
Information About Cisco IOS XR Process Placement
You must be in a user group associated with a task group that includes the proper task IDs. The command
reference guides include the task IDs required for each command. If you suspect user group assignment is
preventing you from using a command, contact your AAA administrator for assistance.
Information About Cisco IOS XR Process Placement
What Is a Process?
To achieve high availability and performance, the Cisco IOS XR software is built on a modular system of
processes. Each process provides specific functionality for the system and runs in a protected memory space
to ensure that problems with one process cannot impact the entire system. Multiple instances of a process can
run on a single node, and multiple threads of execution can run on each process instance.
Under normal operating conditions, processes are managed automatically by the Cisco IOS XR software.
Processes are started, stopped, or restarted as required by the running configuration of the router. In addition,
processes are checkpointed to optimize performance during process restart and automatic switchover.
Process Placement
What Is Process Placement?
Process placement is the assignment of placeable processes to specific locations, such as an installed in the
router.
Placeable processes include all routing processes, such as Open Shortest Path First Protocol (OSPF), Border
Gateway Protocol (BGP), and multicast routing.
Default Placement Policy
In a new system, processes are distributed according to their affinity values among the available nodes and
node pairs in a .
Note
The default process policy that is shipped on the system upon startup is suitable for general purposes.
While customizing is possible, there is no requirement to change the process placement. If you believe
the a change is required, you should work closely with Cisco personnel to ensure that the impact to your
system is contained to just an instance of a process to avoid any undesirable results.
Following is the default placement policy:
Processes have a preference to run on paired nodes (nodes that have an associated standby node).
•
Processes have a preference to remain on their current node. Therefore, processes do not move
•
automatically, unless the unpaired node (or both nodes in a node pair) on which they are running fails.
If the node fails, and there is no standby node, the processes are restarted on a different node.
When a new node pair is added, the following rules apply:
•
The currently running processes are not automatically moved to the new cards.
◦
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
8
Process Placement
Reasons to Change the Default Process Placement
The general preference is for new processes (such as a new ISIS instance) to start on the new node
◦
pair, which contains the most available CPU and memory resources in the system.
Other affinity settings may override the general preference. For example, if the IS-IS process has
◦
a strong affinity to run on the same node where ipv4_io is running, then IS-IS would be started on
that node, and not the new node-pair.
Reasons to Change the Default Process Placement
Although the default process policy that is shipped on the system upon startup is suitable for general purposes,
changes to the router configuration can result in the need for processes to be rebalanced among the available
CPU and memory resources.
When a system is initially booted, the system assumes that all processes use the same amount of memory,
thereby treating each process as equivalent. As the configuration grows, however, the CPU load and memory
requirements of some application processes increase. Centralized applications may need a larger portion of
the resources.
In addition, when a new is added to a system, only new processes or process instances are added to the node.
This could result in some processes with too few resources, while the newer cards are underutilized.
Therefore, as the software configuration changes, or hardware is added, it may become necessary to rebalance
processes among the available in .
Reoptimizing Process Placements
The easiest and most reliable method for users to redistribute processes among the available in is with the
placement reoptimize command.
During router operation, the actual resource usage of each process is collected and compared to the router
configuration and network topology. An ideal configuration for process placement is created and updated in
real time.
To implement this ideal process placement configuration, enter the placement reoptimize command in
EXEC mode. Before the changes are made, the system displays a summary of the predicted changes. You
can either accept the changes or cancel the operation.
See Reoptimizing Process Placement, on page 11 for detailed instructions.
Reconfiguring Process Placements
You can also change the process placement affinities, or preferences, to override the default policies. For
example, you may learn that some processes perform better on the primary node pair, or that some processes
have better high-availability characteristics when running on a paired node (a node with a standby partner).
Other processes might benefit from co-location or by being assigned to nodes far apart from each other.
Note
Consult with your technical support representative before changing the default process placement
configuration. Incorrect configurations can cause system error, poor performance or downtime.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
9
Reconfiguring Process Placements
Recommended Guidelines for Process Placement
The following are a few recommended guidelines for changes to the process placement configuration:
Generally, the process placement feature functions well upon system startup; fine tuning is seldom
•
required.
Use the EXEC mode command placement reoptimize , as described in the Reoptimizing Process
•
Placements , on page 9 to automatically redistribute the processes among the available .
Keep process placement policy changes to a minimum, and always consult technical support personnel
•
before implementation.
Process Placement Based on Memory Consumption
You can change process placements based on memory use of processes. Memory use is expressed in terms
of the memory “footprint of the placeable process. The system attempts to spread the load among the nodes
without exceeding their memory capacity. In addition, the system computes the affinity values to determine
the best placement.
Cisco IOS XR software assumes that every placeable process uses one megabyte of memory.
For detailed instructions, see Setting Memory Consumption Thresholds, on page 12.
Process Placement
Changing Process Affinities
Process placement can also be controlled by changing the affinities, or preferences, of a process or process
group. The following types of process affinities are operator configurable:
affinity location set
•
affinity location type
•
affinity program
•
affinity self
•
affinity location set
This affinity specifies a preference for a process to run on a specific node pair or set of node pairs. A node
pair is either an active and standby pair of nodes [hosted on ], or a single active node on an that does not have
a standby.
affinity location type
This affinity specifies a preference for a process to run on a particular location type. Available location types
are as follows:
• paired— nodes that have an associated standby node
• primary—Primary node
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
10
Process Placement
affinity program
affinity self
How to Configure Cisco IOS XR Process Placement
• current —Current node. A process’s affinity to its current node characterizes its preference to remain
on the same node where possible.
You configure the placement policy to allow certain processes to stay where they are (current ) or move by
specifying the various affinity values. The higher the positive value of an affinity, the stronger the requirement
that the process run at a location, and so on. A low or zero point value indicates a weaker requirement (or no
preference) that a process run at a location.
This affinity specifies a preference for a process to run on the same node as another process, or to run on a
different node than another process. You would want to use this affinity in the case that certain processes
perform better when they are running together on the same node (attract); or on different nodes, apart from
each other (repulse).
This affinity adjusts placement decisions when multiple instances of a process are started. An attract (positive)
affinity indicates a preference to have all instances of a process run on the same node, while a repulse (negative)
affinity indicates a preference to have each instance of a process run on different nodes.
Hierarchical Placement Policy
When you configure placement policies, you must remember that affinities are applied to the software in a
hierarchical way.
Affinities applied to process instances take precedence over affinities applied to a process class. In the following
example, all OSPF instances have a preference to run on the primary of the , but only OSPF instance 10 has
a preference to run on a paired node:
RP/0/RP0/CPU0:router(config)# placement program ospf
RP/0/RP0/CPU0:router(config-place)# affinity location-type primary attract 200
Class affinities take precedence over default process affinities. In the following example, all OSPF instances
have a preference to be placed on unpaired nodes. This overrides the default policy for all processes to prefer
paired nodes.
RP/0/RP0/CPU0:router(config)# placement program ospf
RP/0/RP0/CPU0:router(config-place)# affinity location-type paired repulse 200
How to Configure Cisco IOS XR Process Placement
Reoptimizing Process Placement
This task reoptimizes the placeable processes among the available nodes according to memory and CPU
usage.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
11
Setting Memory Consumption Thresholds
SUMMARY STEPS
1.
2.
DETAILED STEPS
placement reoptimize
Use one of the following commands:
yes
•
no
•
Process Placement
PurposeCommand or Action
Step 1
placement reoptimize
Example:
RP/0/RP0/CPU0:router# placement reoptimize
Step 2
yes
•
no
•
Example:
RP/0/RP0/CPU0:router# yes
Setting Memory Consumption Thresholds
SUMMARY STEPS
show placement policy global
1.
configure
2.
placement memory {maximum | threshold} value
3.
Use one of the following commands:
4.
Displays the predicted changes of the
optimization.
Accepts or rejects the changes.Use one of the following commands:
end
•
commit
•
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
12
Process Placement
DETAILED STEPS
Creating a Location Set Affinity
PurposeCommand or Action
Step 1
Step 2
Step 3
Step 4
Example:
RP/0/RP0/CPU0:router# show placement
policy global
configure
placement memory {maximum | threshold}
value
Example:
RP/0/RP0/CPU0:router(config)# placement
memory maximum 80
end
•
commit
•
Example:
RP/0/RP0/CPU0:router(config-place)# end
or
RP/0/RP0/CPU0:router(config-place)#
commit
Displays the current memory settings.show placement policy global
Use maximum value keyword and argument to set the maximum
percentage of memory that can be used on a node (based on the estimated
memory usage of the processes).
Use the threshold value keyword and argument to define the memory
load level to trigger migration. The system attempts to balance all nodes
at or below the threshold memory percentage. In other words, the system
does not place a process on a node that has exceeded the threshold value,
unless all other nodes have also reached their thresholds (or unless some
other large affinity overrides this consideration).
Saves configuration changes.Use one of the following commands:
When you issue the end command, the system prompts you to
•
commit changes:
Uncommitted changes found, commit them before exiting
(yes/no/cancel)?[cancel]:
Entering yes saves configuration changes to the running
◦
configuration file, exits the configuration session, and returns
the router to EXEC mode.
Entering no exits the configuration session and returns the router
◦
to EXEC mode without committing the configuration changes.
Creating a Location Set Affinity
This task sets the affinity of a placement program (process) to or from node pairs.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
Entering cancel leaves the router in the current configuration
◦
session without exiting or committing the configuration changes.
Use the commit command to save the configuration changes to the
•
running configuration file and remain within the configuration session.
13
Creating a Location Set Affinity
SUMMARY STEPS
DETAILED STEPS
Process Placement
configure
1.
placement program {program [instance instance] | default}
Sets the affinity of a placement program (process) to or from one
of its own instances.
Saves configuration changes.Use one of the following commands:
When you issue the end command, the system prompts you
•
to commit changes:
Uncommitted changes found, commit them before
exiting (yes/no/cancel)?[cancel]:
Entering yes saves configuration changes to the
◦
running configuration file, exits the configuration
session, and returns the router to EXEC mode.
Entering no exits the configuration session and returns
◦
the router to EXEC mode without committing the
configuration changes.
Entering cancel leaves the router in the current
◦
configuration session without exiting or committing the
configuration changes.
Use the commit command to save the configuration changes
•
to the running configuration file and remain within the
configuration session.
Step 5
Step 6
show placement location {node-id | all}
Example:
RP/0/RP0/CPU0:router# show placement
location all
show placement program {program | all}
Example:
RP/0/RP0/CPU0:router# show placement program
bgp
Displays the location of a placement process.
Displays the operational state for each placement program.
Configuration Examples for Process Placement
This section contains examples to view the processes that are placeable in an SDR.
If you believe that a custom reconfiguration of the processes on your system is required, you should work
closely with Cisco personnel to ensure that the impact to your system is contained to just an instance of a
process to avoid any undesirable results.
System Management Configuration Guide for the Cisco NCS 6000 Series Router, Release 5.0.x
19
Additional References
Process Placement
To learn the processes that are placeable, enter the show placement program all command in EXEC mode.
RP/0/RP0/CPU0:router# show placement program all
Mon Aug 18 17:13:15.155 PST DST
If a program is shown as having 'rejected locations' (i.e., locations on which
it cannot be placed), the locations in question can been seen using the "show
placement policy program" command.
If a program has been placed but not yet started, the amount of time elapsed
since the program was placed is shown in the 'waiting to start' field.
Parentheses around the node indicate that the node has not yet fully booted.