cancel
Showing results for 
Search instead for 
Did you mean: 

BE 2010 R3 Ejects Tape Unexpectidly

SPC-Techs
Level 3

 

Server: HP ML350 G6

Processor: E5504

RAM: 4Gb

Windows Server 2008 R2 SP1

HDD RAID Card: Smart Array P410i

Tape Drive RAID Card:  Smart Array P212

Tape Drive: HP LTO Ultrium-5

Backup Exec 2010 R3 all LiveUpdate patches to 19th Jan 2012 (Cannot update anymore because now see “LU1805: LiveUpdate was unable to find any products to update.”)

 

Hello Symantec Community,

I need some advice on an issue. I am not sure if this is just a PEBCAC issue or a genuine problem, but I have not come across this issue before in my use of BE over the last 2-3 years.

I have an issue where upon completion of the “Full – Copy the files” backup method, BE ejects the tape and requests a new over writeable media, even though the tape is not full, and the check box “Eject media after job completes” is clear. The issue however, is that the job continues to run, and eventually hits the “Job has exceeded the maximum allowable runtime” barrier.  

Further to this, if you insert another tape in the drive whilst the job is still running, the data continues to be copied to the new tape, even though we know that when the initial “eject” happens, all the data appears to be getting backed up. We have done a few “disaster recovery” tests and we were happy with the result so far.

This issue just all of a sudden popped up. Everything was going fine, jobs were succeeding, and then it just randomly started.  For the record, I also had this same issue pop up at a similar time on another backup server in our domain. This server is at a remote site and has different hardware (HP DL380 G5 and LTO-4) but after about 2 months it just sort of went away with no intervention from myself (other than live update patches.)

We use BE as basically a file copy method to copy from HDD to LTO-5 and to keep it secure, so we are not doing anything to exceptional. We do full backups, not incremental or differential. Just plain old “Full – Copy the files” and “Overwrite Media.” Even allowing for a verify, the same issue happens.

I have tried formatting all the tapes, creating new media sets, creating a separate ‘test job’ and using that for a while, deleting the original job and recreating it, performing a ‘repair installation’ of BE, and even uninstalling\reinstalling BackupExec. Windows 2008 R2 is fully patched also. Formatting and reloading Windows 2008 R2 is an option (as this is the only role this server does) but I would rather not if it can be avoided.

Some assistance would be greatly appreciated. Screen shots and Job Summary\configurations are hopefully attached, if I did it right!

11 REPLIES 11

pkh
Moderator
Moderator
   VIP    Certified

You should not be using the "Copying file method" to make a copy of your backup.  You should be using a duplicate job.  The best way to duplicate a backup is to use a policy with a backup template and a duplicate template.

As for your problem, after your job, you should click on the tapes in the Media tab and check the tape statistics.  It is probably because the first tape is full.  What makes you think that it is not full?

SPC-Techs
Level 3

Hello pkh,

 

I am not sure what you mean by "I should not be using 'Copy the files' method." To explain a bit more, we use a third party backup application to perform snapshots of our VM's. These VM's are snapshot from their host servers and sent via the network to a share on the backup server. The backup server then uses BE to copy these files to the LTO-5 media. This is more for a 'fire backup' than anything as it means we have an offsite backup. For normal day-to-day we use the third party program and it's snapshots for restores.

As for the the first tape been full, the size of the backup folder is about 300Gb less than what is there. For example, the folder size for the last job (last night) is 1.25Tb. The size of the BE job is 1,281,414,038,579 bytes. Near Enough. The tape hardware is LTO-5 which holds 1.5-1.6-ish-Tb uncompressed so I should be clear for space. And because I have selected "Overwrite Media" when the job begins I was under the impression that BE would just overwrite all, regardless of what is there. BE isnt saying that the tape is full, just to "Insert overwriteable media". Some of the jobs are less than 1Tb (i.e some have been 800-900Gb) and we still have the same issue.

Hope this clears some things up, and I can provide some job logs if needed to help clarify things.

Thanks for the swift reply. 

 

 

yusufusmani
Level 4

After reading the hole issue, going thorugh the JD and snaps, I have found in your job properties that u have

"Eject media after job completes: Yes"

"Enable automatic cancellation for this selection list: Yes"

"Cancel backup job if not completed within: 13 hours"

I not sure about these options, but suspect the issue could be because of the above or one more thing noticed in snap is the "Media Set" which shows "0" hours for Over write Protection and "0" hours for Append Period. media set may also cause problems when the job runtime is more then the Append Period.

you can try 24 hrs or 1 day for both the option may solve the problem here.

Please update the resolution if u manage to achive the target.

community experts please correct me if I am wrong here!

Regards.

pkh
Moderator
Moderator
   VIP    Certified

I thought you are using BE to backup to disk and then copying the files to tape.  This copying is not recommended.  What you are doing should be o.k.

Go to Tools ---> Options ---> Media Management and check that the Overwrite Protection Level is set to either Partial (default) or Full.  Do not set it to None as this turns off overwrite protection.

Also, set your OPP and AP to at least 1 hour.

SPC-Techs
Level 3

Hi PKH,

You may be on to something here.

Because of our backup needs I actually want overwrite protection turned off. As well as backing up to disk, we also backup to the afformentioned remote site (we are lucky enough to have a 1Gbps 'dark fiber' link between sites) so we have no issue with overwriting tapes on weekly basis, i.e Mondays tape gets overwritten every Monday. We have Monday-Tursday tapes and Friday 1-5 tapes. At the moment, media overwrite protection level is set to 'partial' and 'prompt before overwriting imported meida' is also checked. I will set this to none and give it a try. 

We have a public holiday in Australia on the 26th Jan, so I will get back to you guys hopefully on Friday the 27th Jan after a backup job has run at the schedule time.

 

Thanks again for the advice.

pkh
Moderator
Moderator
   VIP    Certified

I don't think you should turn off overwrite protection.  You run the risk of accidentally overwriting your data.  you should set the OPP appropriately so that the media is not protected when it is needed.  Once you set the overwrite protection level to None, it will turn off overwrite protection globally.

In any case, there are a number of cases in the forum in which the user tries to turn of data protection and get funny results.  If you don't want a long OPP, then set the OPP to 1 hour and the overwrite protection level to Partial.

SPC-Techs
Level 3

Hi PKH,

Back to work today, and the job failed again with the same symptoms, but after just reading your post about 'funny results' from turning off OWP this may be why. 

I have changed the OWP to 'Partial - protect only allocated media' but will leave the 'promt before overwriting imported media' unchecked. I have also set the OPP to 1hr as requested.

Today is Friday and we are not back at work untill Monday and don;t have enough time to run a whole job through, so I will let you all know on Monday how it goes.

 

Thanks again for the advice.

SPC-Techs
Level 3

 

Hello Symantec,

 

Ooooookay. Sorry for the late reply but we have been under the pump at work with the start of the new school year, and I have been busy testing solutions with this issue.

The previous advice about Overwrite Protection didn’t make a difference, the job kept failing still. However, we now have successful backups!

Let me explain. When this started happening last year sometime, the first thing I did was click on error support link, and read the advice on the KB article. At the time, it appeared to only list that to fix the error you need to insert a tape, and\or look at SQL or Windows Update service. We had checked those already, and this did not fix the issue. However, when clicking on the KB article link now, there is another: http://www.symantec.com/business/support/index?page=content&id=TECH125395&actp=search&viewlocale=en_...

Now, this is not explicitly the exact scenario of the problem, but is close enough. I  unshared the old location, created another folder, gave it the same share name, and changed the location of our VM snapshot program to backup to this new location. Performed a test backup with one of the VM’s (about 100Gb) and it worked fine. As our snapshot files retention increases, so does the size of the backup, and so far we have had 5 successful backups in a row, with the largest backup 780Gb so far. I will keep an eye on it, just in case the job gets to the old backup size of 1.1-1.3Tb and we start having problems again.   

 

Thanks again for the assistance.

SPC-Techs
Level 3

Hello Symantec,

I have some good and bad news. The good news is that we got about two weeks of solid, successful backups. The bad news is that the problem has returned. I have tried the tape idea suggested in the knowledge base article (where you perform a quick erase and relabel media) but this hasn’t worked. I also tried to perform a long erase and a rename, and no luck.

Obviously we can move the backup location to another share\folder structure again, but it would be quite a hassle if we had to do that every couple of weeks or so. I am also going to try to do a patches update, but I am getting the "LU1805 LiveUpdate was unable to find any products to update" again. I will try to work on getting this resolved so I can perform updates, but some assistance on the above issue would be appreciated.

Thanks.

jedwards2
Not applicable

I just resolved the same LiveUpdate error on our server by doing the following:

 

  1. Uninstall LiveUpdate
  2. Reboot
  3. Insert the install DVD
  4. Navigate to the LiveUpdate installation folder on the disc
    (for us it was "D:\BE\WinNT\Install\LiveUpdate\EN")
  5. Run the LiveUpdate installer directly - on our disc it was LUSETUP.EXE
  6. Open a command prompt 
  7. Navigate to the Backup Exec program folder - we'd used the standard folder:
    "C:\Program Files\Symantec\Backup Exec"
  8. Run the following command:
    BEUpdateOps.exe -AddBE -OptOut

SPC-Techs
Level 3

Hi jedwards2,

Yeah I found a similar solution to the failed updater problem elsewhere which worked, but thanks anyway.

However I am still having this same issue as listed above.

I have now upgraded to BE 2012 and am still having the same problem. I have found so far that it doesnt appear to be the job itself, as the last few days I have been selecting one folder at a time and adding a new one each day. This works because we only backup about 10 folders. Backing up the first 6 or so folders manually or at the jobs runtime seemed to work fine, but then we get the error again. This happens if the size of the folders is 800Gb or 1.3Tb, ie. job size does not have a bearing on result.

 

I am adding another folder to the list and trying it manually today to see result. Hope someone can help out.