Print | Rate this content

HP ProCurve Switch 5400zl Series - Network Connection Drops Every 1 to 2 Minutes

Issue

After upgrading from 13.63 to 14.41 the network is inaccessible to users every 1 to 2 minutes.

The two core switches are in a Virtual Router Redundancy Protocol (VRRP) setup. The primary switch is not responding to ping every 2 minutes intervals. It will start again for a short while and then stop responding.

On connecting to the switch console it reports:

System went down: 10/21/09 20:40:41
Saved crash information:
NMI event HW:IP=0x00f39b70 MSR:0x02029200 LR:0x00f39ae0
cr: 0x88000800 sp:0x02f08158 xer:0x20000000
Task='tDevPollRx' Task ID=0xa9a2000

ROM information:

Build directory: /sw/rom/build/bmrom(t2g)
Build date: Apr 24 2009
Build time: 23:33:11
Build version: K.12.20
Build number: 24648

Boot Profiles:

0. Monitor ROM Console
1. Primary firmware image
2. Secondary firmware image

Select profile (primary):

Booting Primary firmware image...

Decompressing...done.

Solution

The best thing to do is obtain a SHOW TECH ALL if possible and also the CRASH LOGS:

CRASH LOG: You will need a TFTP server to get this binary file. Use the following CLI command: (Use "master" and slots A-F only if 5406zl)
(Hint: You may want to use [UP ARROW] to recall previous commands.)
copy crash-log master tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log a tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log b tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log c tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log d tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log e tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log f tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log g tftp (TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log h tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log I tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log j tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log k tftp <TFTP server's IP addr> <filename> [*ENTER*]
copy crash-log l tftp <TFTP server's IP addr> <filename> [*ENTER*]

Further analysis of the SHOW TECH output revealed that the switch crashes when it is on 14.41 with either the wireless or redundant module (on 01.28) installed and crashes.

No module installed on 14.41 and all ok.

On K.13.63 the wireless module is stable and no crashes.

We also have issues with the wireless module causing HSL error on 14.41 on other modules:

W 10/21/09 23:20:46 00564 ports: port E19 PD Invalid Signature indication.
I 10/21/09 23:20:46 00076 ports: port B18 is now on-line
W 10/21/09 23:21:27 00374 chassis: HSL Fatal F0: SLOT A HSL #7 - HSL status
E78012D2
W 10/21/09 23:21:27 00374 chassis: HSL Fatal F0: SLOT A HSL #8 - HSL status
868012D2
W 10/21/09 23:21:27 00374 chassis: HSL Non-Fatal F0: SLOT A HSL #7 - HSL status
E78012D2
W 10/21/09 23:21:27 00374 chassis: HSL Non-Fatal F0: SLOT A HSL #8 - HSL status
868012D2
I 10/21/09 23:21:27 00077 ports: port A2 is now off-line
I 10/21/09 23:21:27 00077 ports: port A3 is now off-line
I 10/21/09 23:21:27 00077 ports: port A4 is now off-line
I 10/21/09 23:21:27 00077 ports: port A5 is now off-line
I 10/21/09 23:21:27 00077 ports: port A6 is now off-line
I 10/21/09 23:21:27 00077 ports: port A7 is now off-line
I 10/21/09 23:21:27 00077 ports: port A8 is now off-line
I 10/21/09 23:21:27 00077 ports: port A23 is now off-line
I 10/21/09 23:21:27 00077 ports: port A24 is now off-line
W 10/21/09 23:21:28 00274 chassis: (82) Ports A: Blade Crash detected -
Available
I 10/21/09 23:21:34 00375 chassis: Slot A Downloading
I 10/21/09 23:21:35 00376 chassis: Slot A Download Complete
W 10/21/09 23:21:43 00274 chassis: Slot A HSL0 Fatal 71001f2: Task=tDevPollRx
Task ID=0x61fd000 IP=0x235474

I 10/21/09 23:21:47 02631 SNTP: server not found at 10.1.10.11.
I 10/21/09 23:21:53 00422 chassis: Slot A Ready

To obtain the SHOW TECH off the WESM you can connect to it and issue support copy tech-support tftp://xx.xx.xx.xx/tech-support.tar.gz.

The temporary fix here was to regress back to K.13.63. The recommendation was:

  • Upgrade the WESM to release 1.30.

  • If there is no specific reason to upgrade to K.14.41, then it is recommended to use K.13.72 with the WESM code 1.30. On the flip side of that, if you need features available in K.14.41 or above, then we do have installation successfully using this combination as well.

Provide feedback

Please rate the information on this page to help us improve our content. Thank you!