CBS service crash on NW2004s NWDI SP14

[Former Member]

Hi,

We've experienced a couple of crashes of the CBS service after an upgrade from SP09 to SP14.

So far they only relevant entries I found in the log are the following:

#1.5 #0012799E50B600D9000002A4000039F000044E4B44BE7D37#1211985060877#/Applications/CBS/Service##com.sap.tc.cbs.server.rt.cache.src.verifier.impl.LogCacheInconsistency####n/a##dbf719102cc011ddc85c0012799e50b6#Thread[SAPEngine_System_Thread[impl:5]_Group - 67,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Error#1#com.sap.tc.cbs.server.rt.cache.src.verifier.impl.LogCacheInconsistency#Plain###

The source cache is in inconsistent state. (Request ID: 16984, BS: JDI_ZSRMMDM7_D [BSID: 143])#

#1.5 #0012799E50B600D600000003000039F000044E4B4BDB90BF#1211985180218#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.BinarySweeper####n/a##d7bf62d02cc011dd86060012799e50b6#Thread[CBS-Binary-Sweeper 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Info#1#com.sap.tc.cbs.server.rt.impl.BinarySweeper#Plain###Sweep unused binaries away#

....

#1.5 #0012799E50B600EA00000001000039F000044E5EA41FA366#1212068261519#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.Pulsar####n/a##6937a3d02d8411dd9a5f0012799e50b6#Thread[CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Error#1#com.sap.tc.cbs.server.rt.impl.Pulsar#Java### faces a problem time in sequence#2#CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6#1st# #1.5 #0012799E50B600EA00000003000039F000044E5EA41FA4D1#1212068261519#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.Pulsar####n/a##6937a3d02d8411dd9a5f0012799e50b6#Thread[CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Error#1#com.sap.tc.cbs.server.rt.impl.Pulsar#Java### faces the following problem: #2#CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6#java.lang.IllegalStateException: CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6 did not confirm instance activity for more than 180 sec.# #1.5 #0012799E50B600EA00000005000039F000044E5EA41FA713#1212068261520#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.Pulsar####n/a##6937a3d02d8411dd9a5f0012799e50b6#Thread[CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Fatal#1#com.sap.tc.cbs.server.rt.impl.Pulsar#Java###CBS *Instance pulse thread made trial(s) to recover in last sec without success. In order to prevent malicious side effects, this CBS instance must be shut down*.#3#CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6#java.lang.IllegalStateException: CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6 did not confirm instance activity for more than 180 sec.#180#

#1.5 #0012799E50B600EA00000007000039F000044E5EA41FA8D4#1212068261520#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.Pulsar####n/a##6937a3d02d8411dd9a5f0012799e50b6#Thread[CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Fatal#1#com.sap.tc.cbs.server.rt.impl.Pulsar#Java###CBS Instance pulse thread triggers instance shut down procedure.#1#CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6# #1.5 #0012799E50B600EA00000009000039F000044E5EA41FAA0F#1212068261521#/Applications/CBS/Administration##com.sap.tc.cbs.server.rt.impl.DeadNodeDetector####n/a##6937a3d02d8411dd9a5f0012799e50b6#Thread[CBS-Pulsar 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6,10,SAPEngine_System_Thread[impl:5]_Group]##0#0#Info#1#com.sap.tc.cbs.server.rt.impl.DeadNodeDetector#Java### is to be stopped#1#CBS-Deadnode-Detector 37956050/d68b76b0-2cc0-11dd-bf89-0012799e50b6#

Any idea of where the root cause could be ?

Regards

Dagfinn

[Former Member]

Closing as it has not been a problem lately

[Former Member]

I think the key message is

com.sap.tc.cbs.server.rt.cache.src.verifier.impl.LogCacheInconsistency

all other messages after this seem to inply clean up and death.

So it looks like a cache inconsistency inside CBS..... I have scanned everywhere for you with no luck so the approach I would recommend now is to

1. Double check te CBS log and defult.trc for times just beforfe this event.

2. Raise a call with SAP

3. It depends on the frequency of error since update if only a couple of times over months then perhaps this is not so dangerous....

4. see what activity was happening at the times it happened for any correlation.

I would imagine it is a problem with the CBS application rather than the objects in workspaces/buildspaces.

I guess you could initialize compartments in desperation also just to make sure CBS has the bast data possible and highlight any issues that may be lurking that should not be.

[Former Member]

Sorry for the slowness of my reply and thanks for your quick reply

The CBS service is still up and running since the last restart so it is at least relatively stable.

Agree that the cache inconsistency appear to be the root cause and I am now initializing the compartments for the buildspace in question.

Will close this thread once the next transport goes ok in.

Regards

Dagfinn

[deidre_logan]

In your issue "

CBS service crash on NW2004s NWDI SP14"

above you mentioned ".....cache inconsistency appear to be the root cause and I am now initializing the compartments for the buildspace in question." 

how did you Initilize the compartment for the buildspace?

thanks

Deidre Logan