April 26, 2011 at 12:16 pm
OK, Here is my set up:
Primary server ( SQL 2008 Ent +sp1) on a A/P cluster;
Secondary is a stand alone ( SQL 2008 Ent + Sp1) in same domain.
Both OS are Windows Server 2008 Ent.
I set up the LS via wizard on primary server when it's on one node and the backup folder is on the SAN drive ( Added as SHAREd FOLDER in the failover cluster manager). it worked great, once set up, all the jobs can run happily....
BUT...(yes, but, i know you are waiting for it ) when the SQL Server on Primary side failed over to another node. the LSBACKUP job failed with error saying : Executed as user: DOMAIN\SQLACCOUNTID. The process could not be created for step 1 of job 0xB2C9889DC40D0842B56A4B7B78DE4E33 (reason: The system cannot find the path specified). The step failed.
This is the part I don't understand, the DOMAIN\SQLACCOUNTID account has full control over that share folder!
How do you set up this backup share so that LSbackup job will run even when there is a failover in the cluster?
I am going to try the mirroring if this doesn't go well. but I would like to know what I am missing here.
TIA
April 26, 2011 at 12:44 pm
Can you do a manual backup to the backup share after you have failed over? Test with any database to both the actual path (ie c:\backups) and the share (ie \\backupshare\backups). Is the backup_share defined in log_shipping_primary_databases the set as the windows cluster name?
SELECT primary_database,backup_directory,backup_share
FROM msdb.dbo.log_shipping_primary_databases
you may get a little more info as to why it is failing in log_shipping_monitor_history_detail.
SELECT *
FROM msdb.dbo.log_shipping_monitor_history_detail
I would also double-check that this exists on both nodes of your cluster
C:\Program Files\Microsoft SQL Server\90\Tools\Binn\sqllogship.exe
April 26, 2011 at 12:51 pm
well, i should have waited to post this.
Just did some further more research online and found this link
This has nothing to do with the backup share folder, it has to do with the DIFFERENT location for sqllogship.exe on the two nodes.
I copied the ######\tools\binn folder to the one which doesn't have, and it is working now. 🙂
---Thanks to Allan Hirt, wherever he is !!!
April 26, 2011 at 12:54 pm
Oops, apparently i spent too much time on the previous reply and didn't see Bob's reply.
Yes, BOB is right. It is the sqllogship.exe that caused the problem. Thank you for taking time to help on this, Bob.
Appreciate it!:-)
November 21, 2016 at 10:06 pm
Yep we also fixed the similar issue by following 2 steps-
1.Modify the path in the job after failover manually
2.Copy the 110\Tools\Binn\*.* to the single location making the path even
I followed the second solution which seems to be long term and will not break after cluster fail over.
Cheers,
Got an idea..share it !!
DBA_Vishal
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply