Group still running although CFI has been written

[vup]

server:solaris 2.6 /networker 613
client:w2ksp4 /networker 613

Group is running,not terminate. There are two incomplete saveset:F: and index.

In fact, saveset F: and index have been written and I can recover them.

Stop group and restart group look can correct the group status to normal but those incomplete saveset will backup again.

This situation does not happen on one specific client.
The more critical problem is that will cause savegroup
will not run on next schedule.

Any reply will be appreciative.

 

[TDun]

what does mminfo for this client show?

 

[605]

Obviously there has been a timeout problem. Look for the message "attempt
to kill on a remote save" in the daemon.log file. This will prove it. However,
this does not tell you why - most likely it is a network related problem.

Sure you can recover incomplete save sets until it ends.

 

[vup]

TDun,

# mminfo -c ods3 -N F:\\ -t "last week"
volume client date size level name
FUB506S1 ods3 02/11/2004 6492 KB incr F:FUB506S1 ods3 02/12/2004 10 MB incr F:FUB507S1 ods3 02/13/2004 9718 KB incr F:FUB508S1 ods3 02/16/2004 18 MB incr F:FUB508S1 ods3 02/17/2004 17 MB incr F:
Lastest line is the result of manual restart. You can see that looks like imcomplete saveset cause group still running has an index at 02/16/2004 .

 

[vup]

605,

I cannot find those strings in daemon.log . I think they will be in if I restarted this group more delayed time.

 

[605]

Sorry, but this tells nothing. How do you recognize the index backups????

 

[vup]

605,

I am sure that incomplete saveset has been backup in fact.
I can find this saveset in certain volume.

 

[eskolnik]

I have this issue all the time, the savesets show up as completed in the daemon log, but the ack back from the client never comes. I see it was FIN_WAIT2 on the networker server.

ed

Ed Skolnik