Backup Jobs Randomly stop and queue
We have a 2003 server running Veritas BE 9 with the backups going to an external Iomega drive. There is a folder created for each day of the week and each job points to that folder and overwrites it weekly. The jobs usually work most of the time with no problem. Every so often the job runs partway and then is stuck at queued the next morning until it is cancelled. It may then happen for a couple of days in a row, then work ok.
There are 3 drives on this server that are rotated in and out each week so that there is always a set of backups offsite. The BE services are stopped and restarted after the drive is switched so that it is recognized and it is mapped as the same letter each time. It seems that almost every week or two at some point the job gets stuck at queued and folders have to be recreated on the drive and in BE for different days. Is there a way to avoid this so that it will run on its own? The jobs have worked at different points going to the folders on each drive on their own, then every so often it starts getting stuck again. The drives have plenty of free space as well.
Thanks.
There are 3 drives on this server that are rotated in and out each week so that there is always a set of backups offsite. The BE services are stopped and restarted after the drive is switched so that it is recognized and it is mapped as the same letter each time. It seems that almost every week or two at some point the job gets stuck at queued and folders have to be recreated on the drive and in BE for different days. Is there a way to avoid this so that it will run on its own? The jobs have worked at different points going to the folders on each drive on their own, then every so often it starts getting stuck again. The drives have plenty of free space as well.
Thanks.
ASKER
The only error is usually from when I cancel the job, otherwise it just sits there at "queued" and the log has nothing new at that point. The job seems to start ok, then gets stuck like that. I know deleting and recreating the folders in backup exec or deleting the 2 .cfg files in the folders and letting a restart of the services recreate them usually fixes it if the job goes immediately to queued when it tries to start. The services are all still running when it is queued and it lets me cancel it. They are not backing up exchange or any databases. I will try and check the patches.
Maybe it's also just a standard windoze error which could get resolved by regularly rebooting the server. I've just noticed that OS's that run on m$ tend to need regular reboots, so I'd schedule one every weekend...
ASKER
Deleting the job in BE (each day has its own job) then deleting the backup to disk folder for that day and recreating the folder and the job allowed the job to run successfully today. The last 2 days it ran and then got stuck just over halfway through and queued. This is the 3rd time we have had to recreate the jobs/folders due to this and it has fixed it temporarily each time. Could switching out the drives each week be causing this even if it is mapping as the same letter?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thanks I'll give that try.
ASKER
Deleting the folder and recreating the job seems to work each time, like if I do it at the start of the week when they change the drives. I wish there was a way to not have to do that each time but it has worked so far.
ASKER
It only seems to happen with Backup Exec with drives that are switched out weekly or daily. Even though the drives all have the same M-F folders on them and map as the same letter, it will queue the job and not write to it until the folders are deleted and the job is recreated, then it will do it again a few weeks later. That is the only fix I have found.
Is the backup disk full when this happens? Have you set the jobs to "overwrite"? If you have it set to "append" and the disk is full, BE wil wait for an empty disk or one with space on it. Also make sure the backups are set so they can be overwritten at the time you need to. If the retension period is set too long, earlier backups won't be overwritten by BE as it is still marked as no overwriteable yet.
ASKER
No there is room on the drive. I have had 2 support calls so far with Symantec/Veritas and they can't figure it out as of yet, the job runs fine it just gets to a random point and cancels itself, sometimes saying that the hardware if offline even though the drive shows up and is working properly. I will post anything they tell me that fixes it.
ASKER
Veritas informed me that the BE software does not support using external drives with the backup to disk folders and swapping out the drives. It can be used as a removable backup to disk folder but then it will overwrite the job each time and won't save other jobs on there. It took the 3rd support call about this for them to tell me that. For now we are going to use the removable backup to disk folder and swap out the drive each day or go to the Windows backup program so we can save more than one day on there. I guess we can leave it as abandoned on this question since they do not appear to have an answer.
Thanks.
Thanks.
Don't abandon a Q but close it. To do that follow the instructions in the link I posted earlier.
As for backing up to removable disks, you could upgrade to BE 10, which should support that.
As for backing up to removable disks, you could upgrade to BE 10, which should support that.
Is BE updated with all current servicepacks and patches?