SAN shared option (SSO) - 6 jobs try to run, but only have 4 tape drives!

[Support_User]

I have managed media 2 servers sharing a 4-drive library.  The Central Admin server pushes 3 jobs to each media server (so 6 jobs in total).  This means that each server can run 2 jobs, while the 3rd job sits queued, on the managed media server, waiting approx 12 hours for a tape drive to become free, even though we have other managed media servers, with non-shared (SCSI) tape drives that are often free after 2 hours.

Argh, it appears the SSO servers don't communicate quickly enough to realise that a tape drive has changed from available to in-use, thus causing my problem.

Anyone seen this, fixed this (I wish!), or had similar setup and not had this issue?

Thanks.

All servers BE2010 R2 on Win 2008 R2.  CASO server is SSO master, and doesn't run any jobs itself.  1 x fibre zones per BE server (so 3 in total)  Each fibre zone includes all tape drives.  Have considered having a 4 x fibre zones per BE server (i.e 12 in total, each zone would include 1 BE server and 1 tape drive), but have tried that before and didn't seem to make any difference).

[Larry_Fine]

Do all 6 jobs launch at once?  Might you be able to hold 2 of the jobs back a bit so all the device usage is known before they get dispatched?  In other words, launch 4 jobs to 4 drives, then later launch the other two jobs and let it hunt for availabel drives?

[Support_User]

Hi Larry, good suggestion, we've looked at that, but we'd like to avoid the administration of having to set availability times on selection lists (or having to have multiple policies) - i.e. we want Symantec to fix what has been an issues since 12.x or earlier.  Problem is, I don't think anyone from Symantec has EVER actually used SSO in a production environment, and they only seem to fix issues that they find themselves, they ignore almost all customer feedback and consistenly treat all issues as if the causes is external to Backup Exec.  I'm not holding my breath.