Diagnostic error and KAL-AP

Question

Hello @all!
On weekend we saw following error in our Cat6509 (12.2(33)SXI) for our ACE-Module  A2(2.3)
switch#sh diagnostic events module 1
Diagnostic events (storage for 500 events, 500 events recorded)
Number of events matching above criteria = 8
Event Type (ET): I - Info, W - Warning, E - Error
Time Stamp         ET [Card] Event Message
08/15 02:35:04.367 E  [1]    diag_get_bus_swt_mode(): Error receiving SCP resp
                             onse forgetting the switching mode
08/15 02:35:09.375 E  [1]    diag_asic_read_reg16[1/1/42]: set_get_asic_regs()
                              failed (get)
08/15 02:35:09.383 E  [1]    queryHyperionSynched[1]: diag_asic_read_reg16 fai
                             led
08/15 02:35:09.387 E  [1]    test_asic_sync[1]: queryHyperionSynched failed
08/15 02:35:10.419 E  [1]    TestAsicSync Failed
08/15 02:35:10.431 E  [1]    diag_pf_const2_test_has_failed[1/-1]: reg_invoke_
                             diag_get_test_attribute_by_name failed
08/15 02:40:19.559 I  [1]    Diagnostics Passed
08/16 12:14:40.460 I  [1]    TestAsicSync Passed
You see, there is an error with the ASIC on the ACE-Module. Because of this the switch was shutting down interface Ten1/1 for about 20sec and made a reset of the module. First question is, if the failure is an hardware-issue?
Because of the reset, we had a failover with the redundant module in the other switch. In front of both we have a GSS which uses "KAL-AP by VIP". We use is for several VIPs and they are still working after the failover. Only for 2 VIPs, the ACE was sending a load of 0. In the config we found a fragment of an old configurations for "KAL-AP by Tag". Configuration looked like this
domain domain_KALAP-BUMS
  add-object class-map cmap-vip_BUMS
Before you ask, the configuration on the GSS was "KAL-AP by VIP" and it was working fine before the failover. After the failover the new active module was sending a load of 0 to the GSS. Is this a bug? Did anyone had some similiar error?
Would be happy if someone can help us understanding.
Kind regards,
Marko

Answer 1

Hi Marko,
For this type of issue we tend to replace the module. Please raise a TAC case in consideration of RMA. Provide the following command output from switch chassis in the TAC case:
1. show-tech
2. syslog from switch at time of trouble
3. remote command switch show scp status
4. remote command switch show scp accounting
5. remote command switch show eobc
As for the your KAL-AP question, your description is a bit vague.
It is not clear to me what the chain of events were, nor if all is ok now?
If the two VIPs which reported load 0 still do, and it is in error, then this must be investigated.
Hope this helps.