Re: TCP errors
1998-10-01 16:31:11
Look into your COMMTIMEOUT value (from dsmadmc do a "q opt") and at
your IDLETIMEOUT we were suffering tons of these and in one
environment had to crank the settings up to:
CommTimeOut 7,800 IdleTimeOut 120
to get around sessions being dropped, now the problem you describe
below is self correcting but others might not be.
IBM, what ever happened to the little ADSM heart beat / I'm still here
just still working on things ? ? ?
______________________________ Reply Separator _________________________________
Subject: TCP errors
Author: Tom.Fluker (Tom.Fluker AT viasystems DOT com) at unix,mime
Date: 10/1/98 12:33 PM
Has anybody seen the following errors in their dsmerror.log file:
TcpRead: Zero byte buffer read.
sessRecvVerb: Error -50 from call to 'readRtn'.
TcpFlush: Error 9 sending data on Tcp/Ip socket 4294967295.
sessRecvVerb: Error -50 from call to 'readRtn'.
ANS1809E Session is lost; initializing session reopen procedure.
This happens periodically on both normal ADSM backups as well as
SQL-BACKTRACK backups (the BACKTRACK sessions don't restart like the normal
backup sessions).
Server: AIX 4.1.5 running ADSM 3.1.1.3
Client: Solaris 2.5.1 running ADSM 3.1.0.3
Any ideas??
Tom Fluker
|
|
|