[c-nsp] Cisco 10008 %IDMGR-3-INVALID_ID: bad id in id_get

[Rin]

Hi group, 

 

I currently facing a problem on Cisco 10008 with following logging output: 

Dec  9 02:03:56.669: %IDMGR-3-INVALID_ID: bad id in id_get (Out of IDs!)
(id: 0x5514A038)

-Traceback= 40796DA0 407972E4 407B1090 40710290 41ECB96C 414DDAF0 41EE5770
41EE59A8 41ECFC68 41ECFFE4 41ED01BC

Dec  9 02:03:57.341: %IDMGR-3-INVALID_ID: bad id in id_get (Out of IDs!)
(id: 0xD4AA415C)

-Traceback= 40796DA0 407972E4 407B1090 40710290 41ECB96C 414DDAF0 41EE5770
41EE59A8 41ECFC68 41ECFFE4 41ED01BC

Dec  9 02:03:57.353: %IDMGR-3-INVALID_ID: bad id in id_get (Out of IDs!)
(id: 0x5514A038)

-Traceback= 40796DA0 407972E4 407B1090 40710290 41ECB96C 414DDAF0 41EE5770
41EE59A8 41ECFC68 41ECFFE4 41ED01BC

 

This causes high CPU utilization on the device: 

BRAS#sho process cpu

CPU utilization for five seconds: 92%/80%; one minute: 77%; five minutes:
75%

 PID Runtime(ms)   Invoked      uSecs   5Sec   1Min   5Min TTY Process 

   1         440      8825         49  0.00%  0.00%  0.00%   0 Chunk Manager


   2      410708   2921477        140  0.07%  0.01%  0.00%   0 Load Meter


   3           0         3          0  0.00%  0.00%  0.00%   0 OBFL Cfg
Dispatc 

   4     3912208  67345954         58  0.00%  0.01%  0.00%   0 C10K Card
Event  

   5           0        51          0  0.00%  0.00%  0.00%   0
Retransmission o 

   6           0         2          0  0.00%  0.00%  0.00%   0 IPC ISSU
Dispatc 

   7    41321232   3022609      13670  0.00%  0.53%  0.63%   0 Check heaps


   8        1204      6488        185  0.00%  0.00%  0.00%   0 Pool Manager


   9           0         2          0  0.00%  0.00%  0.00%   0 Timers


  10           0         2          0  0.00%  0.00%  0.00%   0 Serial
Backgroun 

  11        4388   2911625          1  0.00%  0.00%  0.00%   0
ALARM_TRIGGER_SC

 

It seems that this problem is bug CSCsl28246 with IOS 12.2(33)SB, but the
workaround from Cisco is not clear for me: 

CSCsl28246 

Symptoms: Not able to bring up more than 32768 TC Sessions and Out of IDs
AAA trace back message is displayed. 

Conditions: This symptom occurs under TC sessions. 

Impact: Traceback preventing scale of ISG PPP Traffic Class. Scalability
issue. 

Trigger: While running ISG sessions with PPPoL2TP LAC/LNS on Cisco 10000,
unable to bring up more than 32768 TC sessions because of the following Out
of IDs AAA trace back message: 

Nov 13 11:00:56.696 EST: %IDMGR-3-INVALID_ID: bad id in id_get (Out of IDs!)


AAA is allocating only 1024*32 = 32768 IDs. Not able to bring up any more
sessions because of accounting flow id allocation failure. 

Workaround: Increase the number to number of sessions/flows required on the
platform. 

Can anyone elaborate more on this workaround? Thanks a lot. 

FYI, I'm running IOS c10k3-p11-mz.122-33.SB2.bin on this platform.

Rin