ACE failure

Unanswered Question
Apr 12th, 2010


we have 2 ACE module with the next version :

  loader:    Version 12.2[120]
  system:    Version A2(1.5a) [build 3.0(0)A2(1.5a) adbuild_19:28:51-2009/07/31_/auto/adbu-rel4/rel_a2_1_5_throttle/REL_3_0_0_A2_1_5
  system image file: [LCP] disk0:Cisco-c6ace-t1k9-mz.A2_1_5a.bin
  installed license: ACE-VIRT-020 ACE-SEC-LIC-K9

  Cisco ACE (slot: 1)
  cpu info:
    number of cpu(s): 2
    cpu type: SiByte
    cpu: 0, model: SiByte SB1 V0.2, speed: 700 MHz
    cpu: 1, model: SiByte SB1 V0.2, speed: 700 MHz
  memory info:
    total: 955660 kB, free: 231892 kB
    shared: 0 kB, buffers: 4784 kB, cached 0 kB
  cf info:
    filesystem: /dev/cf
    total: 1000512 kB, used: 420736 kB, available: 579776 kB

last boot reason:  Unknown
configuration register:  0x1
slb-ace01-02 kernel uptime is 3 days 15 hours 43 minute(s) 30 second(s)

those module are in failover mode?

since few days an issue appear on the standby module, we can't access the module with remote access (radius) and on the console access we just have a cli with exit command allowed and an error message :

Process did not respond within the expected timeframe, please try again.

Error getting profile for user admin_0

Error: Role-domain checking init failed

we have decided to reboot it and so failover the only context master on this module, after this every think come back as normal.

But the master module has the same issue actually, it is the master for all context.

it seams that we are locked out all management functions. there are no core dump or any log that explain why.

is anybody have an idea please???


I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Sean Merrow Tue, 04/13/2010 - 09:06


This can happen when one or more of the CP processes are too busy to handle the connection.  There were a couple of enhancements added in A2(1.6) software to allow the login to work.  I would recommend an upgrade to A2(2.3) on your modules to get those, and many more, fixes.



antony_cisco Wed, 04/14/2010 - 02:21


thank you for our answer.

But is it possible to know there are busy processes?

I guess that the CP processor will be overload and then in the monitoring we could see this. But we haven't any cpu changed when the ACE have the failure.

by the way, all issues with management features confirm an error with the control plane.

we are waiting for a version more recommanded by cisco tac.


Sean Merrow Wed, 04/14/2010 - 10:16


Once in this state, it is probably impossible to tell what process is busy as the CLI will not allow you to debug the issue.  It is typically due to cfgmgr and/or aclmerged.  There have been some enhancments to allow the CLI to still work.  I would recommned going to A2(2.3).  Also, A2(2.4) has also just been posted.



This Discussion