S.P 8.1.0 core dump with ANR7823S Internal error UNSUP_ARGTYPE

[Nicke]

Hi there

I have a new backup system using S.P / TSM v8.1.0 (Extended Edition) in a kind of Semi-Production status.
OS is Redhat Linux 7.1 (x86-64)

This morning a noticed that the D/R Replication target server instance had core-dumped and DB2 proceeses where still runing and No error / warning shown in .../db2dump/db2diag or tsminst notify.

The crash occurred when another TSM v8.1 instance ran node replication to this TSM server.


cat dsmserv.err:
02/20/2017 08:00:47 ANR7824S Server operation terminated.
02/20/2017 08:00:47 ANR7823S Internal error UNSUP_ARGTYPE detected.
02/20/2017 08:00:47 ANR9999D 0x0000000112b0c8 pkAbort
02/20/2017 08:00:47 ANR9999D 0x0000000107728a PrepDispArgs
02/20/2017 08:00:47 ANR9999D 0x000000010781e1 outFFDCLog
02/20/2017 08:00:47 ANR9999D 0x00000000c38e0f tbFlushInsCache
02/20/2017 08:00:47 ANR9999D 0x0000000095fbe6 imUpdateInventory
02/20/2017 08:00:47 ANR9999D 0x00000000996fc9 imPrepareTxn
02/20/2017 08:00:47 ANR9999D 0x0000000100c489 tmEndX
02/20/2017 08:00:47 ANR9999D 0x00000000ddc5a5 SmEndVbTxn
02/20/2017 08:00:47 ANR9999D 0x00000000f18dbf SmReplServerSession
02/20/2017 08:00:47 ANR9999D 0x00000000dbfc94 DoReplServer
02/20/2017 08:00:47 ANR9999D 0x00000000dc1ded smExecuteSession
02/20/2017 08:00:47 ANR9999D 0x00000001142f6e psSessionThread
02/20/2017 08:00:47 ANR9999D 0x0000000112bd12 StartThread
02/20/2017 08:00:47 ANR9999D 0x007ffff76abdf5 *UNKNOWN*
02/20/2017 08:00:47 ANR9999D 0x007ffff35231ad *UNKNOWN*
02/20/2017 08:00:47 ANR7820S Server thread 2 (tid 140737107760896) terminated in response to program abort.
...

I had no problem restarting this TSM instance and "audit container" showed 0 failed containers.
I'll fix an IBM PMR in a while but first checking if there is any new S.P 8.1 server fix patch but I cant find any.

//Nicke

 

[Nicke]

Hi there

I have a new backup system using S.P / TSM v8.1.0 (Extended Edition) in a kind of Semi-Production status.
OS is Redhat Linux 7.1 (x86-64)

This morning a noticed that the D/R Replication target server instance had core-dumped and DB2 proceeses where still runing and No error / warning shown in .../db2dump/db2diag or tsminst notify.

The crash occurred when another TSM v8.1 instance ran node replication to this TSM server.


cat dsmserv.err:
02/20/2017 08:00:47 ANR7824S Server operation terminated.
02/20/2017 08:00:47 ANR7823S Internal error UNSUP_ARGTYPE detected.
02/20/2017 08:00:47 ANR9999D 0x0000000112b0c8 pkAbort
02/20/2017 08:00:47 ANR9999D 0x0000000107728a PrepDispArgs
02/20/2017 08:00:47 ANR9999D 0x000000010781e1 outFFDCLog
02/20/2017 08:00:47 ANR9999D 0x00000000c38e0f tbFlushInsCache
02/20/2017 08:00:47 ANR9999D 0x0000000095fbe6 imUpdateInventory
02/20/2017 08:00:47 ANR9999D 0x00000000996fc9 imPrepareTxn
02/20/2017 08:00:47 ANR9999D 0x0000000100c489 tmEndX
02/20/2017 08:00:47 ANR9999D 0x00000000ddc5a5 SmEndVbTxn
02/20/2017 08:00:47 ANR9999D 0x00000000f18dbf SmReplServerSession
02/20/2017 08:00:47 ANR9999D 0x00000000dbfc94 DoReplServer
02/20/2017 08:00:47 ANR9999D 0x00000000dc1ded smExecuteSession
02/20/2017 08:00:47 ANR9999D 0x00000001142f6e psSessionThread
02/20/2017 08:00:47 ANR9999D 0x0000000112bd12 StartThread
02/20/2017 08:00:47 ANR9999D 0x007ffff76abdf5 *UNKNOWN*
02/20/2017 08:00:47 ANR9999D 0x007ffff35231ad *UNKNOWN*
02/20/2017 08:00:47 ANR7820S Server thread 2 (tid 140737107760896) terminated in response to program abort.
...

I had no problem restarting this TSM instance and "audit container" showed 0 failed containers.
I'll fix an IBM PMR in a while but first checking if there is any new S.P 8.1 server fix patch but I cant find any.

//Nicke

FYI:
This morning the same type of errors are seen in the actlog and in dsmffdc.log during the "node replication" from the other "front-end" TSM server and as it turns out some of it's nodes have active node replication from yet another TSM 7.1.7 server ... This is the main problem but still TSM should not crash as it did yesterday.

//Nicke