Exchange 2003 duplication from disk to tape taking two hours longer after dupe is done.

[wuchgwan_dunow]

Hello:

I'm running BE 2010 R3 on a 2008 R2 SP1 64-bit box. All the latest patches and updates installed. RAWS pushed to all appropriate servers.

My problem:

My Exchange 2003 full backups to disk are about 808 gig and take about 4 to 5 hours to finish. S'all good.

The duplications to tape run, seem to finish in about 3 hours, 40 minutes, but then continue to run on what seems like nothing for another 2 hours.  The byte count reaches the same size as the backup. The job details fields go blank e.g. "Source, Destination, Current Directory, Current File." The file count climbs very slowly. But the bite count doesn't. It's at the max, because it's the same size as the backup to disk. And it sits like that for 2 hours.

So, you see, it seems like it's done. But it takes an extra two hours to .. cycle through this process.

I've searched on this one and called tech support, but this is a tough one. And it's not so important to keep someone on the phone forever to try to figure out what's causing it. If anyone knows what's going on here, would be much appreciated.

-- Sass

[Ken_Putnam]

Is there an active alert to remove the tape from the drive?  If so, the job clock will continue to click

You can set an AutoAnswer of "OK" to end the job if that is indeed the case

[wuchgwan_dunow]

Thanks, Ken.  I'll look for that next time. I haven't seen any alerts come up at that point. And we have an auto-loader. So, it should just put the tape back in its slot when it's done, so far as I am experienced.

I have noticed that since my upgrade to R3, some of my alerts and especially my media intervention alerts have not been popping up consistently. e.g. I'll import tapes into the bot, and the media intervention alert has always come up asking for tape replacement for the slots I chose. Since the upgrade to R3, this only happens sometimes. I can't figure a pattern on it. Sometimes a reboot of the server will fix it. The media intervention alert does pop up in the event logs, and eventually the import will reach its time-out and run. But it's a new PITA.

That may be a rabbit-hole conversation, though. As I am pretty certain this 2-hour latency issue with the dupes has been around long before the R3 upgrade.