ISCSI Target - The disk is offline

[NZ16]

I'm running the new Backup Exec 2012 with all the current patches.  I have disk to disk backups scheduled.  The odd thing is sometimes I get the error the disk is offline and the error contains this: This is typically caused by the folder becoming inaccessible due to it being deleted, renamed, or unshared. It may also be caused by a disk full condition. The folder state has been set to offline. Please attend to this condition. Additional detail may be found in the Event Viewer - Application Log.  

The ISCSI target is always available, anyone have any solutions or seen this before?

[CraigV]

Hi,

Make sure the iSCSI target is persistent. You can also get hold of the b2dtest.exe tool and make sure there are no issues with your setup by running the tool against the drive.

Thanks!

[NZ16]

I have a QNAP device configured as the ISCSI target and it is marked as persistent already.  I run the b2dtest.exe tool and it fails, but I think it is because I don't have the proper path. If I run one off backups, they almost always work fine, but the scheduled ones have issues.

[CraigV]

...mmm...check the TNs below. They only go up to BE 2010, but if this problem has cropped up in BE 2012 and you can't resolve it as per the recommendations, head on over to the Known Issues Section (https://www-secure.symantec.com/connect/backup-and-recovery/issues) and check to see if this has been listed as a known problem. If it has, vote it up...alternatively, if you have support, you might also want to open a query around this...

http://www.symantec.com/business/support/index?page=content&id=TECH124232

http://www.symantec.com/business/support/index?page=content&id=TECH54598

Have you also considered removing the B2D folders from BE (don't delete!!!), and then readding them back in? Have you tried to recreate a scheduled job to see if this runs? If it does, then it could point to some sort of corruption in the original job...

[Matt12345]

I had this issue awhile back (with BE 10 maybe??).

The BE services would start before the iSCSI initiator came up. The fix was to one of the BE services dependent on the MS iSCSI initiator - so BE would wait for the iSCSI initiator before attempting to start.

I forgot which one I made it dependent upon - maybe the BE Engine or BE Media services..

[NZ16]

The disk always says online when I am in the console - so no problem with running a backup "now" job.  I don't know why the disk would go offline at any time.  The services aren't starting or restarting for the jobs to run so I can't see how that would relate.

[CraigV]

...and the NAS and switch logs show no errors?

[NZ16]

No errors...

[CraigV]

...might be worthwhile seeing if there are any MS patches for the iSCSI initiator. Download them and install them.

Unpresent your disks from the server, and then represent them and run a backup.

[BEI]

While I have acces to the B2D drive from windows explorer using Iscsi, BE 2012 tells me it's offline.

Any Ideas.

Also How to re run the job I want and not that BE allows me 2.

I would like to be in control of what jobs I can manualy re-run, not the next job, because the next job is incremental and I want to run the full one because the incremental fail, because full one failed the first time, or the next scheduled job which is totaly different.

I want to cry, seriously, I'm at the end of my wits here.

[CraigV]

...check the Known Issues section to see if this issue is listed there...

https://www-secure.symantec.com/connect/backup-and-recovery/issues

It might also be worth your while opening a support case with Symantec. If, or when, they help you, post back here and close it off.

Thanks!

[BEI]

I have spent all day working on Backup issues.

I have now found how to run individual jobs:)

I have also recreated the job and it appers to be working, I feel sorry for people with many serves.

Regards

[teiva-boy]

I would recommend if you have the chance to stress test the iSCSI system using either the Symantec B2D test tool, or some other tool like an ATTO benchmark or IOMeter.  

Basically run a repeating test to see if you can recreate the disk disconnecting and receiving an error.  Give it a bunch of sequential writes at multiple streams and see what happens.  

Oh and I hope you have MPIO setup on multiple paths or are using 10GbE.  Without MPIO, speed is dismal with 1GbE iSCSI.

[NZ16]

I think I'll try a few more things, then will have to call support and open a case if I can't get it working.  Thanks for the suggestions.

[BEI · Accepted solution]

Hi,

Just letting you know that the problem with our iSCSI disconnections to the Dedupe store etc... on Dell3600i is now resolved.

We have moved our Vmware vCentre to a new server away from the storage managment and that seems to have resolved our issues.

regards