15-minute delay between volumes?

[Charlie_Earle]

Hello all,

I've got a group of 5 servers I am backing up to LTO tape nightly using BE 10d (5629, SP1). The backups run fine, but the total backup time is extrordinarily long due to an unexplained "pause" between each volume.

On examination of the logs, it appears that there is an approximately 15 minute lapse between the time the backup of one volume ends and the next begins. For example, on the local BE server, the log reports that the backup of C: ends successfully at 10:15:26 PM, but the backup of D: does not start until 10:30:09 PM. This 15-minute time seems to be consistent between all volumes, and between servers.

I am absolutely confident that this is not a network or remote agent problem, because the same time delay occurs even between backing up the local C: and D: volumes on the local media server, with no network involved. Also, I have ensured that Display Progress is turned off.

Please, any advice or ideas would be very much appreciated.

Thanks!

[Steve_Brown_3]

Hi Charlie,

I've seen this before when using the Advanced Open File Option and selecting the option to snapshot each volume seperately. Get these done at once, and you'll be on a roll.

However, I suggest you create seperate jobs for each server, as it isolates things in case of any failure.

Hope that helps

Steve

[Deepali_Badave]

Hello,

Run the sgmon before the backup job.

How to create SGMON.LOG
<<>>

NOTE : If we do not receive your reply within two business days, this post would be marked assumed answered and would be moved to answered questions pool.

[Charlie_Earle]

Hi Steve,

Thanks for the help! Funny thing is, I'm not using AOFO on the jobs at all - maybe I should be? I haven't purchased AOFO as an option, but I'm still running in "trial mode" before I open any of my license packs (just in case...I like to be cautious that way), so trying out AOFO is possible at this point. I definitely agree that creating seperate jobs for each server is a good idea, and I'll probably do that at some point along the way - but right now it doesn't seem that it will help with this probelm much, since the pause is happening between volumes on each server as well as between servers. I did set up a "test" job on just the media server itself just to be sure to isolate the problem from network issues - same results.

I'm going to use SGMON, as Deepali suggested, on tonight's job and report back with the results. Hopefully some deep logging will shed some light on what the hangup is.

Thanks for the suggestions!

Charlie

[Charlie_Earle]

Problem solved - I'm happy to report that the problem was completely resolved by simply running a database repair using BEUTIL - doing this cut down the time to backup 130 GB over 5 servers from 8 hours to 1.5 hours - just a tad more reasonable.

I'm not sure why the database from a fresh install had problems to begin with, but logging a few test jobs with SGMON showed that there were occasional database errors, along with several "Device going to sleep for 900000 msecs" entries - which explains the 15-minute pauses, but with no real explanation why. On a hunch, I ran a database repair and compact, and *voila*. Each volume is now beginning backup immediately after the backup of the last.

It troubles me some that I don't have an exact answer, but I can't argue with the results. Only hope this helps someone else with the same issue.

Thanks to all for the help!

Charlie