Not sure when it cropped up, but I'm on 4.2.3.0 on Solaris and AIX and we
have to recycle the server before 65k session count or we get hundreds of
9999D's per hour.
There is also another issue with the summary table where certain client
versions can't update the summary table on certain fields which produces
9999D's also.
01/28/03 11:26:29 ANR9999D smnode.c(18972): ThreadId<40> Session
exiting
has
no affinityId cluster
01/28/03 11:26:29 (40) Context report
01/28/03 11:26:29 (40) Thread AcceptorThread (33) is a parent thread
related
to: 40
01/28/03 11:26:29 (33) Generating TM Context Report:
(struct=tmTxnDesc)
(slots=256)
01/28/03 11:26:29 (33) *** no transactions found ***
01/28/03 11:26:29 (33) Generating Database Transaction Table Context:
01/28/03 11:26:29 (33) *** no transactions found ***
01/28/03 11:26:29 (33) Generating SM Context Report:
01/28/03 11:26:29 (33) *** no sessions found ***
01/28/03 11:26:29 (33) Generating AS Vol Context Report:
01/28/03 11:26:29 (33) No mounted (or mount in progress) volumes.
01/28/03 11:26:29 (33) Generating ssSession Context Report:
01/28/03 11:26:29 (33) No storage service sessions active.
01/28/03 11:26:29 (33) Generating ssOpenSeg Context Report:
01/28/03 11:26:29 (33) No storage service segments found.
01/28/03 11:26:29 (33) Generating BF Copy Control Context Report:
01/28/03 11:26:29 (33) No global copy control blocks.
01/28/03 11:26:29
01/28/03 11:26:29 (40) SessionThread : ANR9999D calling thread
01/28/03 11:26:29 (40) Generating TM Context Report:
(struct=tmTxnDesc)
(slots=256)
01/28/03 11:26:29 (40) *** no transactions found ***
01/28/03 11:26:29 (40) Generating Database Transaction Table Context:
01/28/03 11:26:29 (40) *** no transactions found ***
01/28/03 11:26:29 (40) Generating SM Context Report:
01/28/03 11:26:29 (40) Session 66145: Type=Node, Id=DLEAG
01/28/03 11:26:29 (40) Platform=SUN SOLARIS, NodeId=210,
Owner=oracle
01/28/03 11:26:29 (40) SessType=4, Index=6, TermReason=0
01/28/03 11:26:29 (40) RecvWaitTime=0.000 (samples=0)
01/28/03 11:26:29 (40) Backup Objects ( bytes ) Inserted: 0 ( 0.0
)
01/28/03 11:26:29 (40) Backup Objects ( bytes ) Restored: 0 ( 0.0
)
01/28/03 11:26:29 (40) Archive Objects ( bytes ) Inserted: 0 ( 0.0
)
01/28/03 11:26:29 (40) Archive Objects ( bytes ) Retrieved: 0 ( 0.0
)
01/28/03 11:26:29 (40) Last Verb ( SignOff ), Last Verb State (
Recv )
01/28/03 11:26:29 (40) Generating AS Vol Context Report:
01/28/03 11:26:29 (40) No mounted (or mount in progress) volumes.
01/28/03 11:26:29 (40) Generating ssSession Context Report:
01/28/03 11:26:29 (40) No storage service sessions active.
01/28/03 11:26:30 (40) Generating ssOpenSeg Context Report:
01/28/03 11:26:30 (40) No storage service segments found.
01/28/03 11:26:30 (40) Generating BF Copy Control Context Report:
01/28/03 11:26:30 (40) No global copy control blocks.
01/28/03 11:26:30
01/28/03 11:26:30 (40) End Context report
-----Original Message-----
From: Cook, Dwight E [mailto:DWIGHT.E.COOK AT SAIC DOT COM]
Sent: Tuesday, January 28, 2003 10:34 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: At what level does the 4.2.x server have the session count
fix?
what platform ?
when did it break ?
I'm at 4.2.2.0 on AIX and I get into the millions on a regular basis...even
10's of millions as I seem to remember.
Sess
Number
-------
1,354,1
76
1,365,0
47
1,369,9
14
Dwight
-----Original Message-----
From: Jolliff, Dale [mailto:xjolliff AT TI DOT COM]
Sent: Tuesday, January 28, 2003 10:30 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: At what level does the 4.2.x server have the session count fix?
In 4.2.3.0, after the session count gets to 65K, the session number can no
longer be stored in the summary table resulting in an ANR9999D message.
(Can anyone say maxint?)
I'm pretty sure a fix is out... I hope.
|
