June 17, 2008 at 1:44 am
Strangely It failed again at 6:45AM. (Online Backup starts at 5:00AM).
Will try to get some more info and will get back to you.
Appreciate your patience.
June 17, 2008 at 1:49 am
Yes please do.
June 18, 2008 at 1:37 am
Hello,
just confirmed the error. It started going Out of Sync at 6:15AM and got the first Event Error at 7:00AM.
So it really seems to be something with the Online Backup.
Strangely I've checked the Backup software and the scripts are not configured to do Log Cleaning.
I already did the Schedule test to stop before Backup and restart Backup, the same errors came out.
Any more ideas? Is there any extra info you need?
Thanks once again.
June 18, 2008 at 3:11 am
if a log backup is happening, or a truncate log, it will be recorded in your errorlog. go through that to see if you can find evidence of a log backup outside of your logshipping plan.
---------------------------------------------------------------------
June 18, 2008 at 7:19 am
No errors or Truncate on the Logs.
Tomorrow I will have a final report from the Backup Team, but so far I can't seem to find a logical reason behind this Out of Sync on the LSRestore SQL Agent.
June 18, 2008 at 7:31 am
be sure there are no other jobs which would truncate the transaction log. I found a similar situation on one server I worked on, where someone must have been concerned about the log file size so had set up a TSQL job to issue a backup LOG with TRUNCATE_ONLY statement. (why?). This of course will mess up your log backup chain as soon as the statement is executed. I did not discover this until I set up log shipping from this server to a standby server.
Make sure no job is using backup log with TRUNCATE_ONLY, or that some part of a maintenance plan is not doing that.
The probability of survival is inversely proportional to the angle of arrival.
June 18, 2008 at 7:48 am
One other thing... make sure "WITH NO_LOG" is not being used somewhere in a job statement (or maintenance plan). That will break your log chain also.
The probability of survival is inversely proportional to the angle of arrival.
June 20, 2008 at 2:06 am
Hello,
Confirmed the problem with Legato.
Did a new LS on a server that are not on the backup schedule and no errors so far.
Will check the Legato Script but this might be hard to solve.
June 23, 2008 at 6:51 am
Hello again,
I've created a new LS but with different restore times. So instead of 15 minutes it's now on 5 minutes. Might be useless but nothing like testing.
I got a bit of the backup script.
Maybe someone could give some hints?
Thank you.
online )
for SID in $sql_sid
do
log_clean $SID $servername1
done
for SID in $sql_sid2
do
log_clean $SID $servername2
done
clean_up $servername1
clean_up $servername2
savegrp -v $BACKUPNAME
savegrp -v $BACKUPNAME2
;;
log_clean ()
{
set -x
cd \\\\$2\\backup_${1}
find ./* -mtime +0 -type f -depth -print | xargs rm -f
return 0
}
June 23, 2008 at 7:15 am
well that looks like it is removing files from a backup directory, but cannot say what files fromminfo supplied. I recommend getting together with admin responsible for legato andgoing thru the legato sql agent set up
---------------------------------------------------------------------
June 30, 2008 at 1:25 am
Did a new test so I could confirm that the problem was with Log Truncate.
Started a new LS, let it run for a couple of hours and queryed this command:
BACKUP LOG database WITH TRUNCATE_ONLY
GO
I got the Out of Sync again but not on the Schedule I was expecting.
This time it gave an error on the Backup Job instead of the Restore Job.
Is there another way to Truncate a Log?
Thank you.
July 3, 2008 at 12:51 pm
Some questions:
What is the transaction log backup interval?
How long are the transaction logs maintained on the primary server?
What is the copy interval?
How long are the copied trans logs maintained on the secondary server?
July 4, 2008 at 5:22 am
Tlog backups are done every 15 minutes in the LS.
I've set the Tlogs to be deleted after 7 days.
July 16, 2008 at 8:18 am
I have had this situation happen to me. The transaction logs have been deleted from the Primary Server before being applied to the Secondary Server and now the logs are out of sync, how can this issue be resolved?
Thanks
July 16, 2008 at 8:23 am
aliciakw (7/16/2008)
I have had this situation happen to me. The transaction logs have been deleted from the Primary Server before being applied to the Secondary Server and now the logs are out of sync, how can this issue be resolved?Thanks
unless you get tran log backups from a tape backup, you have to start logshipping again from a full backup. Thats it.
---------------------------------------------------------------------
Viewing 15 posts - 16 through 30 (of 32 total)
You must be logged in to reply to this topic. Login to reply