Skip to main content

Backup fails / reschedules itself continuously

Thread needs solution

This morning (at about 1AM) some backups that have been running automatically for several months started behaving strangely. 

On this machine I have three backups, one for each partition, set to backup once a day (at 9:00 PM, 9:30 PM, and 10:00 PM).  They ran properly last night at the appropriate times.  Then I noticed high activity this morning on the machine and that's when I found that they were all running again and again.  Each time saying they failed and would be re-run in 9 minutes.

The system is Windows 10 Pro 64-bit.  The target for the backups is a share from another Windows 10 Pro machine.  A test connection executes properly.  And there are files being created on the target machine in the correct directories.  If I watch one execute, it seems to run to completion and then says it failed at the end.

Any suggestions would be appreciated.  The first thing that would be interesting to know is how I can find out why it thinks the backup failed.

 

Thanks.

David

0 Users found this helpful

David, welcome to these user forums.

Please download a copy of the Log File Viewer app (link in my signature below) and use this to look at the log files that ATIH will be creating for your failed / repeating backups.

Next, please select your backup tasks, select to look at the Options for these, go to the Advanced page tab and check the settings for Error Handling.

The fact that they are rescheduling after 9 minutes suggests that you have the default Error Handling settings - please increase the delay time before any retry of a failed task to a period in hours instead of the default of seconds. This will help stop the constant rescheduling which in turn can start filling your storage drive.

Steve,

Thanks for the quick response.  After I sent the original post I continued looking around the forum and found the log viewer.  I used it and find that a random selection of service log files (including the one attached to this post) seem to say the operation was successful.  Yet the console says it failed and queues it to re-run in nine minutes.  Any thoughts?

As for the Error Handling option, it is set to 5 attemps with a delay of 30 seconds.  I don't believe I ever have toucvhed this setting.  So, I can change it, but this would not seem to be where the magic nine minute delay is coming from.  But I will increase the settings.  It just seems a bit odd to me that it says 5 seconds but does 9 minutes.

Again, thanks for your help.

David

 

 

Attachment Size
381462-132259.log 3.31 KB

Slava, from the Acronis team has posted in other threads that the error handling is indeed **EDIT*** not working correctly in v6571, but is scheduled to be resolved in the final release of 2017.  Steve has helped validate this bug in another thread as well.   Unfortunately, it seems to disregard the time set there :(

There's nothing else in that particular service log - do any of the ones before it show anything else?  I'm wondering if it is trying to run a consolidation and although the backup is successful, it is having trouble with the cleanup.  I ran into this problem in 2017 beta 1, but it logs the consolidation error as well.  I have downloaded 2017 beta 2 (3070) and the consolidation seems to be working on network shares in this version.  If interested, you could join the 2017 beta and test out the latest version with it and see if it helps.  The beta is fully functioning and provides 3 installs. 

 

Sorry for the edit, I meant to say "not" instead of "now" above. 

All the service logs I looked at (a random sampling of the many that were being created) seemed similar ... showing successful completion.

It has now been running for two nightly backup cycles without the problem.  I noticed two things that I changed (breaking my rule of diagnosis of only changing one thing at a time).  I saw that memory usage on the machine was very high, so I moved a video surveillance application to another machine.  I don't really suspect that this was the issue.  But it was related to the fact that I would periodically see one of the Acronis tasks start to use large amounts of memory for a few minutes.  That made me wonder if it was trying to get more and failing ... but that should have logged some sort of error.

So the other thing I found was that the monitor.log file was very large.  It was over 1GB and growing; too large to view easily.  So I deleted it.  That seems to have corrected the problem.  I then was able to look at the new monitor.log file that got created and found that there were frequent messages saying that the sync agent was not running.  Indeed, it was not, so I changed its service setting to start automatically.  (I am not using sync functionality.)

 

David, the monitor.log file is often very large and I also have needed to delete this at times in order to make it readable.  The sync agent messages are also very common in that log.