October 19, 2015 at 7:19 am
Hello --
We are running 2008R2 64-bit server on a Windows 2012 R2 Standard 64-bit distribution. Our schedule includes seven full backups of twenty-three databases that start simultaneously on Friday at 7:00 in the evening. Each job includes from one to five, depending on the size, databases. The backup scripts are from Ola Hallengren's MaintenanceSolution.sql file. This schedule has worked well for some time.
This past weekend, all seven backup jobs failed with six reporting failures on Sunday at 7:30 in the evening, and the last reporting a failure at 7:00 this morning. A check of the server log files, indicated the following entries:
Write on "T:\<destination folder>\<database name>.bak" failed: 170(The requested resource is in use.)
Msg 3013, Level 16, State 1, Server <database instance>, Line 1
BACKUP DATABASE is terminating abnormally.
Outcome: Failed
Duration: 1.23:25:46
Date and time: 2015-10-18 19:25:46
I checked the destination drive for the backups, and there was not a shortage of space for the output files. Additionally, the transaction log backups go to the same location, and there has not been a problem with them. The differential backup job is scheduled to run this evening, and I will be monitoring its run.
What would cause the above problem to occur, and how can I take measures to correct and prevent it from happening again?
October 20, 2015 at 12:53 pm
Is it run from a SQL Scheduled Job ?? I find the "Log to Table" option under the step's "Advanced" tab often has more useful information ... if you have it selected ahead of time.
October 20, 2015 at 1:09 pm
Hello --
Thank-you for your reply. The jobs in question are run as SQL Scheduled jobs, and the "Log to Table" option is set to Yes in the command syntax listed in the Step tab.
I have spoken with one of my coleagues, and a TSM backup has been set up for the server. Could this be a factor?
October 20, 2015 at 2:39 pm
TSM ?
October 20, 2015 at 11:25 pm
TSM could be the issue if it is running the backup at the same time as your scheduled job as part of a disk backup, although I never found it an issue; as I recall there is an option to backup files without locking; I had a separate TSM task set up eventually to deal with database files.
If there is a separate SQL backup being taken then it could be breaking the LSN chain, although this has only affected Differential backups in my experience.
I think you need to look at the TSM schedules and file options and work around them or change them to times more suited to your SQL backup needs.
...
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply