Guest

Cisco BPX 8600 Series Switches

The cnfnodeparm Command Parameters for IGX 8400/BPX 8600 Switches

Document ID: 40825



Contents

Introduction
Prerequisites
      Requirements
      Components Used
      Conventions
The cnfnodeparm Command
      The cnfnodeparm Command Template
      The cnfnodeparm Command Parameter Descriptions
Related Information

Introduction

This document describes the cnfnodeparm command node parameters that affect network performance for Cisco IGX 8400 and BPX 8600 series switches. You must log on with a SuperUser or Service level username to configure the parameters.

Note: Although node parameters are assigned to the local node, they affect the entire network. When using these parameters in large networks, you must carefully consider the affects of your actions to protect performance and reliability.

Prerequisites

Requirements

Readers of this document should be knowledgeable of the following:

  • Strong knowledge of IGX/BPX platforms and networks.

Components Used

The information in this document is based on the software and hardware versions below.

  • IGX 9.2.x switch software

  • This document can generally be used to understand the parameters on other switch software versions of IGX and BPX.

The information presented in this document was created from devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If you are working in a live network, ensure that you understand the potential impact of any command before using it.

Conventions

For more information on document conventions, see the Cisco Technical Tips Conventions.

The cnfnodeparm Command

To make parameter changes, use the cnfnodeparm command, followed by the parameter number, then the desired setting.

The cnfnodeparm Command Template

The following cnfnodeparm command template shows the parameter default values, except for the recommended values which are shown in bold:

1  Update Initial Delay   [ 5000] (D)   16 CC Redundancy Cnfged   [    Y] (Y/N) 
2  Update Per-Node Delay  [30000] (D)   17 MT3 Pass Through Relay [    Y] (Y/N) 
3  Comm-Break Test Delay  [30000] (D)   18 Nw Pkt Tx Rate (pps)   [  500] (D) 
4  Comm-Break Test Offset [   10] (D)   19 TFTP Memory (x 10KB)   [  210] (D) 
5  Network Timeout Period [ 1700] (D)   20 Standby Update Timer   [   10] (D) 
6  Network Inter-p Period [ 4000] (D)   21 Stby Updts Per Pass    [   50] (D) 
7  NW Sliding Window Size [    1] (D)   22 Gateway ID Timer       [   30] (D) 
8  Num Normal Timeouts    [    7] (D)   23 GLCON Alloc Timer      [   30] (D) 
9  Num Inter-p Timeouts   [    3] (D)   24 Comm Fail Delay        [   60] (D) 
10 Num Satellite Timeouts [    6] (D)   25 Nw Hdlr Timer (msec)   [  100] (D) 
11 Num Blind Timeouts     [    4] (D)   26 CBUS Delay (msec)      [   20] (D) 
12 Num CB Msg Timeouts    [    2] (D)   27 SNMP Event logging     [    Y] (Y/N) 
13 Comm Fail Interval     [10000] (D)   28 TFTP Grant Delay (sec) [    1] (D) 
14 Comm Fail Multiplier   [    3] (D)   29 TFTP ACK Timeout (sec) [   10] (D) 
15 Temperature Threshold  [   50] (D)   30 TFTP Write Retries     [    3] (D) 

31 FRP Link Status Alarm  [    Y] (Y/N) 46 Modem Polling Timer    [    1] (D) 
32 Job Lock Timeout       [ 1000] (D)   47 Verify CBA for non-FRP [    Y] (Y/N) 
33 Max Via LCONs          [20000] (D)   48 Send Abit Early        [    N] (Y/N) 
34 Max Blind Segment Size [ 3570] (D)   49 Abit Tmr Multiplier    [    0] (D) 
35 Max XmtMemBlks per NIB [ 3000] (D)   50 Abit Tmr Granularity   [    3] (D) 
36 Max Mem on Stby Q (%)  [   33] (D) 
37 Trk Cell Rtng Restrict [    Y] (Y/N) 
38 Stat Config Proc Cnt   [ 1000] (D) 
39 Stat Config Proc Delay [ 2000] (D) 
40 Enable Degraded Mode   [    Y] (Y/N) 
41 Enable Rrt on Comm Fail[    Y] (Y/N) 
42 Auto Switch on Degrade [    Y] (Y/N) 
43 Max Degraded Aborts    [  100] (D) 
44 Max Htls Rebuild Count [  100] (D) 
45 Htls Counter Rest Time [ 1000] (D)

The cnfnodeparm Command Parameter Descriptions

This section describes the cnfnodeparm command parameters and options:

  1. Update Initial Delay

    This value is multiplied by the number of nodes and gives the delay time, in milliseconds, before initiating the sending of any conditional updates to remote nodes after a switchcc or network join command is executed.

  2. Update Per-Node Delay

    The time, in milliseconds, between initiating conditional updates from one node to the next after a switchcc or network join command is executed.

  3. Comm-Break Test Delay

    The interval, in seconds, between Comm Break tests to any node. Increasing the Comm-Break clearance time helps reduce the fan-in effect on receiving conditional update messages (which can be quite large).

    For small networks, where the amount and size of internodal messages are relatively small, decreasing the Comm-Break test delay parameter to 10 or 15 seconds helps to achieve faster recovery of network traffic.

    It is not recommend to configure this parameter lower than 10 seconds at any time for any reason in a larger network due to the possibility of network flood conditions. A normal setting for this parameter in a large network may be set as low as 10 seconds; however, if flood messages occur, the value should be increased to 15 seconds or greater (toward the default value of 30 seconds).

  4. Comm-Break Test Offset

    The factor between the number of Comm Break Test failures and the number of successes with which to declare a node in the Comm Break state.

  5. Network Timeout Period

    The length of time before retrying an unacknowledged network message. If the Network Packet Transmit Rate (NW Pkt Tx Rate) is reduced, you should increase this value to reduce the possibility of message timeouts.

  6. Network Inter-p Period

    Obsolete for switch software 9.2.x and later.

  7. NW Sliding Window Size

    The number of CC messages that a node can simultaneously transmit to the network—you must always set this value to 1.

  8. Num Normal Timeouts

    The number of unsuccessful attempts at sending a network message which triggers the declaration of a Comm-Break with a particular node.

    Total message timeout is Num Normal Timeouts multiplied by the Network Timeout Period (by default, this is 11.9 seconds - 1700 * 7 = 11.9).

    Each message timeout results in an unreachable node, feeder, or standby controller card.

  9. Num Inter-p Timeouts

    Obsolete for switch software 9.2.x and later.

  10. Num Satellite Timeouts

    The maximum number of timeouts for network messages transmitted over a path containing satellite links.

  11. Num Blind Timeouts

    Maximum number of timeouts for blind network message transmission.

  12. Num CB Msg Timeouts

    The number of Comm Break timeouts and retransmissions before the node declares a Comm Break. One successful ACK message clears a Comm Break.

  13. Comm Fail Interval

    The minimum time, in milliseconds, allocated for Comm Fail testing of all trunks on the local node.

  14. Comm Fail Multiplier

    The number of Comm Fail intervals to skip for available lines.

  15. Temperature Threshold

    If the cabinet temperature exceeds this value (defined in degrees Celsius), an alarm is declared.

  16. CC Redundancy Cnfged

    If a node has dual processors and you enable (Y) this parameter, the active processor sends updates to the standby processor.

    If you disable (N) this parameter, a node is unable to perform a switchcc command either manually or automatically.

  17. MT3 Pass Through Relay

    Obsolete for switch software 9.2.x and later.

  18. NW Pkt Tx Rate

    The maximum rate (defined in FastPackets per second) at which network messages are transmitted from this node. In general, network message traffic should not exceed 10 to 15% of the trunk capacity to avoid flooding the trunk.

  19. TFTP Memory (x10Kb)

    A multiplier which the processor uses to automatically calculate the amount of memory partitioned from the statistics memory region (STATRGN) for Trivial File Transfer Protocol (TFTP) files and TFTP/USER statistics.

    caution Caution: Do not change the value displayed.

    This multiplier changes depending upon the processor:

    • 32Mb uses a multiplier of 61

    • 64Mb uses a multiplier of 210

    • 128Mb uses a multiplier of 132

    If you set the multiplier too high, too much memory is allocated for these tasks—robbing memory from other tasks. If you set the multiplier too low, statistics are lost.

  20. Standby Update Timer

    The length of time, in milliseconds, between occurrences of sending standby updates from the active processor to the standby processor.

  21. Stby Updates per Pass

    Number of updates sent to the standby processor per pass.

  22. Gateway ID Timer

    This parameter only applies to structured topologies.

  23. GLCON Alloc Timer

    This parameter only applies to structured topologies.

  24. Comm Fail Delay

    Number of milliseconds for which to wait to start detecting Comm Fail after a switchcc command or network join is executed.

  25. NW Hndlr Timer (msec)

    The amount of time (defined in milliseconds) with which to wait to send messages to, or receive from, a remote node.

  26. Cbus Delay (msec)

    Default processing delay (defined in milliseconds) per card slot used to throttle sending CBUS (control) messages to a card.

  27. SNMP Event Logging

    Enables logging of Simple Network Management Protocol (SNMP) events to an external printer that is attached to the AUX port of a node. Without the attached printer, this parameter is meaningless.

  28. TFTP Grant Delay (sec)

    Delay, in seconds, between processing successive Trivial File Transfer Protocol (TFTP) grants. This parameter is set through Cisco WAN Manager and is a display-only entry on this screen.

  29. TFTP ACK Timeout (sec)

    Maximum time, in seconds, for which to wait for Trivial File Transfer Protocol (TFTP) ACK message from the server. This parameter is set through Cisco WAN Manager and is a display-only entry on this screen.

  30. TFTP Write Retries

    Maximum number of Trivial File Transfer Protocol (TFTP) transmission retries before declaring a timeout error. This parameter is set by Cisco WAN Manager and is a display-only entry on this screen.

  31. FRP Link Status Alarm

    Determines whether a signaling failure on a network-to-network interface Frame Relay Pad card/Frame Relay Module (NNI FRP/FRM) port causes a major alarm.

  32. Job Lock Timeout

    Time that a job will wait for the user interface (UI) lock before aborting the command.

  33. Max Via LCONS

    Maximum number of local connections that can use this node as a Transit node. The maximum values are:

    • IPX = 5000

    • IGX = 20000

    • BPX = 50000

  34. Max Blind Segment Size

    Maximum size of a segment of a blind message.

  35. Max XmtMemBlks per NIB

    Maximum size of transmit queue to any one node. If the queue grows larger than this, the queue is flushed and a Comm-Break is declared to that node.

  36. Max Mem on Stby Q (%)

    Maximum percentage of memory that can be used for updates on the standby controller card queue before the queue is flushed and updates are restarted.

  37. Trk Cell Routing Restrict

    If enabled (Y), the node is permitted to route ATM cell based traffic over Universal Switching Module (UXM) trunks. If disabled (N), the node does not route any ATM cell based traffic over UXM trunks.

  38. Stat Config Proc Cnt

    Maximum number of statistics to process per iteration.

  39. Stat Config Proc Delay

    Delay between iterations in processing statistics.

  40. Enable Degraded Mode

    If enabled, and an abort occurs when a standby controller card is not available (or ready), the node enters the degraded mode rather than aborting and rebuilding completely.

  41. Enable Rrt on Comm Fail

    If enabled, Comm Fail on a trunk causes connections on that trunk to be rerouted.

  42. Auto Switch on Degrade

    In the degraded mode, once the standby processor is ready (in a Standby state), a graceful switchcc command is performed.

  43. Max Degraded Aborts

    Maximum number of aborts that can occur while the node is in degraded mode before the node performs a complete rebuild (which is disruptive to service).

  44. Max Htls Rebuild Count

    If degraded mode is enabled, this is the number of hitless rebuilds the active processor performs (within a given time period) before it enters the degraded mode.

    If degraded mode is not enabled, this is the number of hitless rebuilds that the active processor performs (within a given time period) before performing a full rebuild.

    If you set this value to 255, the processor performs an infinite number of hitless rebuilds.

    Note: Hitless rebuild and degraded mode are only entered if there is no standby processor card available.

  45. Htls Counter Reset Time

    Indicates the time, in hours, in which the number (determined by parameter 46) of hitless rebuilds can occur.

    With the default values, 100 hitless rebuilds can occur within a 1000 hour time period before the node performs a full rebuild or enters the degraded mode. The 101st hitless rebuild within the time period causes a full rebuild or causes the node to enter the degraded mode.

  46. Modem Polling Timer

    The modem polling timer (defined in seconds) depends on the number of voice cards.

    The processor polls CiscoWorks2000 Voice Manager (CVM) or Universal Voice Module (UVM) cards to determine if a modem upgrade is needed (the 2100 Hz tone that the modem uses to identify itself). If there are a large number of voice cards in a node, you should set this value higher to help reduce CPU cycles.

  47. Verify CBA for non-FRP

    Cisco recommends that you set this value to Y (yes) when Universal Switch Module (UXM) cards are present in a node.

    When enabled, this parameter forces each card to verify the cell bus address to ensure the packet belongs to them—this avoids overlapping cell bus addresses. Enabling this parameter also causes rerouting of non-Frame Relay Pad (non-FRP) card and ATM connections. Make this change in a service window and allow time for the network to settle between changes to subsequent nodes.

    If disabled, a card may accept data that is not intended for them. Once the packet is opened, that misdirected packet is discarded. A symptom would be connections with intermittent connectivity problems.

  48. Send Abit Early

    If a connection is derouted, enabling this parameter causes the BPX to send Abit notification after a configurable amount of time—prior to the connection reporting as FAILED, at which point Abit=0 is sent, provided that a Local Management Interface (LMI) is configured at the endpoints.

  49. Abit Tmr Multiplier

    The multiplier times the granularity (determined by parameter 50) equals the configured delay between when a connection is derouted and Abit is sent. This parameter is only used when parameter 48 is set to Y (yes).

  50. Abit Tmr Granularity

    The time, in seconds (in the range from 3 to 255), used with parameter 49, the Abit Tmr Multiplier.


Related Information



Updated: Oct 04, 2005 Document ID: 40825