Job Engine stops daily and backup fails

[Filip_Notebaert]

Since a couple of days now, the daily backup fails with error "The network connection to the Backup Exec Remote Agent has been lost. Please check for network errors.". This happens every time when backing up the Exchange Information Store, which is located on the local server. There are no remote servers in the network.
I am using MS SBS 2003 SP1, Backup Exec 9.1 Rev. 4691 and patched with Service Pack 4a.
The Job Engine is set to auto restart if it should stop.
Also, the bengine.exe crashes. This is the error in the eventviewer: "Faulting application bengine.exe, version 9.1.4691.54, faulting module bengine.exe, version 9.1.4691.54, fault address 0x000aa215."

Does anyone know how to fix this issue ?
Any help would be much appreciated.

Best wishes for 2006,
Filip

[Filip_Notebaert]

Any news on this please.

Thanks in advance !!

[Rucha_Abhyankar]

Hello Filip,

Under what account is the Remote agent for Windows servers service running? It should run under Local System Account.


What all have you selected for the backup?

Are you also backing up SQL in the same job? If so then please split the backup job into Exchange data and SQL data.

For test purposes just select some mailboxes for backup and then verify.

Hope this helps!


==================


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.

[Filip_Notebaert]

Hi Rucha,

the service is indeed running under the Local System Account.
I am backing up SQL and Exchange in the same job, but it has been working like that for over two years.
Anyway, could you please post a link on how to backup Exchange and SQL in seperate jobs, but on the SAME tape.

I will test the backup of individual mailboxes, verify, and post the result here.

Filip

[Filip_Notebaert]

Rucha,

I just checked, and there is NO way to create seperate backupjobs for SQL and Exchange. All options for SQL selections are "greyed-out", I cannot change anything.

As for Exchange, I only have options FULL, INCREMENTAL or DIFFERENTIAL backup. I can not deselect it.

Please provide other information.

Thanks,
Filip

[Filip_Notebaert]

Please cancel last post. I am currently running a backup of only Exchange, and no SQL.

[Filip_Notebaert]

Hi,

the backup of only Exchange completed successfully, so this is certainly not a corrupt mail issue. I think it might have something to do with the bengine.exe crashes.
To what could that be related ?

Thanks for your reply,

Filip

[Filip_Notebaert]

Hi,

yesterday, I configured 2 jobs to run. First job backups the Exchange, with the option overwrite. Second job backups the userdata, with the option append to media.

First job runs fine; second job ends with the same error, mentioned above, allthough this time not when backing the Exchange Information Store.

This is getting really annoying, since no good backup has run since the 12th of December.

I need a solution for this, my client is starting to lose credibility in the product, and me for recommending it.

I have been using BE for a long time at lots of clients, and had some issues before, but then poeple at Veritas were always very helpful. I seem to be missing that now.

Please, I need a solution fast.

Thanks,

Filip

[Amruta_Bhide]

Hello,
- Do you mean User data under the shadow copy components?
- Did you configure the jobs to start around the same time?

- Can you please check if your PORT 10000 is open and listening?

- If you are running ONLY Local Jobs, can you try as a test to let all the services under LSA and then run a JOB.

******************************************************************
*****************************************************************

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.


Thanks.

[Filip_Notebaert]

Hi,

- Shadow copy components were never included in the backup. Just simple userdata.
- The Exchange backup starts at 9 pm, takes about 45 minutes and the second job starts at 11:30 pm.
- Port 10000 is not open now, but I will open it for test purposes. Should that be TCP or UDP ?
- I will set services to run under LSA for testing and will post the result Monday after the weekend.

Filip

[Filip_Notebaert]

Hello,

after doing all the changes mentioned above, the SAME problem STILL OCCURS.

I really need a solution, help me out here, or have your developers make a patch or something. It is clear that more and more users are having this same problem.

Thanks.

[Asma_Tamboli]

Hi Filip,


Could you please update the JET components for microsoft?

Also, perform a database repair with the help of the following document:

http://support.veritas.com/docs/265180

- is a drwatson log generated when the job engine crashes?

Could you please check out the Event logs for any relevant event IDs when the job engine stops.

Also, check out if there are any active alerts which are waiting to be responded to.

Hope it helps!


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.

[Filip_Notebaert]

Hi Asma,

the JET components were already up to date, that is not the problem.
Database was repaired, but this was no solution. Still the same problem.
As you can read in my first post, the eventviewer shows the following error : "Faulting application bengine.exe, version 9.1.4691.54, faulting module bengine.exe, version 9.1.4691.54, fault address 0x000aa215."
There are NO active alerts awaiting to be responded to.

Please provide another solution.

Thanks,

Filip

[Trevor_Walker]

We have the same problem with 3 of our clients running Backup Exec 9.1. It is driving us mad.
The backups have been occuring no problem for 2-3 years and now they are all exhibiting the same problem:

The network connection to the Backup Exec Remote Agent has been lost. Please check for network errors.

and

Faulting application bengine.exe, version 9.1.4691.54, faulting module bengine.exe, version 9.1.4691.54, fault address 0x000aa215.



We have applied the latest service pack 4a, defragged the drives, chkdsk /f the drives as suggested. However for this to be happening at the same time on 3 different client servers - suggests that there is a conflict caused either by a windows update or 3rd party software.

Please can someone help!

[Michael_Zirbes]

I am having an identical problem with one of my clients. Backups have een fine for 2 years and starting in December failures started to occur. The Job Engine service would stop during the backup of the Exchange mailboxes.

I am wondering if it was a Microsoft patch that was applied. Since the Backup exec 9.1 was not at sp4a when the problem stsrted and I was unable to apply the 4s patch.

Anyway I removed backup exec from the Windows 2000 server and removed the remote agents from the lient servers and performed a compete reinstall using the latest build of 9.1 and applied SP4a. The problem remains.

My next test will be to diable the autocancelation of the backup if it exceeds 8 hours run time to see if that corrects the issue.

[Michael_Zirbes]

I received a good backup last night. The only changes made were that I turned off the auto cancelation and I made the job a hair smaller by removing a directory. I will post again if the problem returns.

[Jack_Ping]

Is there any new updates on this problem?
I have the exact problem recently, I started a thread two weeks ago, but got no solutions. It seems that Veritas has no cure on this.

[Filip_Notebaert]

Hi Jack,

after long investigation of my own, I discovered that actually, the tape was full, and instead of giving me a Media Alert, the job engine would simply crash, resulting in a backup failure, saying that the connection to the remote agent has been lost ...

Really confusing, and it had set me off on a totally wrong course.

Even though the compressed tape capacity should be 72 Gb, the tape was full at about 45 Gb. And that for the reason that quite a lot pdf documents and JPG files were being backed up, and those files are not compressible anymore.

So try a backup with less data and see what happens. It sure worked for me.

Good luck

Filip

[Michael_Czajkow]

I keep getting the same problem from time to time.
Once the customer's backup reaches some level bengine.exe crashes.

I figured out that this is related to tape capacity. The compressed capacity i.e. 160GB but the BE says tape is full after backing up some 110-115GB.

This is what I found on HP website:
question
Why can't I get 40 GB of data on a 150m tape with the 20/40 GB DDS 8 Cassette Autoloader?

answer
You will typically be able to store 40 GB on a 150m tape; that is to say you will usually achieve a 2:1 compression ratio. However, the compression ratio obtained depends very much on the nature of the data you are storing. Data files, which have already been compressed by software will not compress very well and may actually increase in size. Examples are files processed by PKZIP on PCs and the compress utility on UNIX - the popular graphics formats .GIF, JPEG and MPEG are also already compressed. Some files, such as text and empty databases, compress very efficiently. A compression ratio of 2:1 is typically obtainable with most commercial data.

This is regarding the particular HP drive model but generally it applies to all drives. If you work out some solution let us all know.

Thanks,

[Trent_Greenawal]

My eyes hurt from reading all of this about everyone having the same problem. Well I am here to tell you this error still occurs for us, and our tapes are not out of room. We used to back up almost 38gig on a 40/80 tape for over a year. After some file manipulation, we have now been down to 25 +/- gigs for a few months with no major errors and are now receiving this same error each day that I log into our backup server. Any other solutions veritas?

Thanks,
Trent