MSL Library Windows 2012 R2 MOVE_MEDIUM Timeout

[linkusnet]

I am getting timeout errors when I try to load a tape using Backup Exec.  It's been running on a Windows 2k server and BE 12 for years.  Works great.  I have a a new server running Windows 2012 R2 and it takes the drive and library offline.

I'm running tracer from Symantec to log the commands.  On the 2012 server, after 37 seconds I get a 

ASPI  = SCSI_COMPLETED_WITH_ERROR

HBA  = SELECTION_TIMEOUT

On the old server, after 46 seconds  I get 

SCSISTAT_GOOD

STATUS_SUCCESS

I've powered down everything 5 times.  I've tried HP drivers and also the stock Microsoft drivers.  After 2 days of searching HP and Symantec I need some help.

[pkh]

What is the exact model of the library and tape drive?  Check the HCL below to make sure that it is supported.

http://www.symantec.com/docs/TECH214803

Update the library and the tape drive to the latest firmware

Ensure that the library and tape drive is recognised and functioning correctly in the Windows Device Manager. The library should appear as an Unknown Medium Changer with a Microsoft driver and the tape drive should have a Symantec driver.

Run the HP LT&T utility against the tape drive. Make sure you select the write test and that you have stopped all the BE services beforehand.

[linkusnet]

It's on the HCL.  MSL 2024 with LTO-3 Fiber Channel.

Firmware is updated.

Shows in Windows correctly.

Changer shows Unknown Medium Changer and also tried the HP drivers.

Ran LTT and no problems. 

As I said the unit works great on BE 12.5.

[pkh]

If you are running the HP Insight software, uninstall it.

What is the HBA that you are running?  If it support RAID, then check this document to see that it is in the approved list.

http://www.symantec.com/docs/TECH70907

[VJware]

Is BE 2014 updated with SP1 ?

This KB may help  - http://www.symantec.com/business/support/index?page=content&id=TECH5027

[Larry_Fine]

Can you post your tracer .bin file?

A MOVE_MEDIUM command timeout is much longer than 37 or 46 seconds.  I think it is about 10 minutes.

I think you are getting errors on your FC HBA.  Are there errors in your Windows event logs?

I assume that your new server has a new/different HBA than the old server?  I would focus my attention on the HBA driver and firmware, more than the tape & library side.

Here is what a command looks like in tracer when BE has a timeout (the command took longer to execute than BE was expecting):

ASPI Status                             SCSI_COMPLETED_WITH_ERROR
Target Status                           SCSISTAT_GOOD
HBA Status                              SELECTION_TIMEOUT

A good MOVE_MEDIUM command looks like this:

ASPI Status                             SCSI_COMPLETED_WITHOUT_ERROR
Target Status                           SCSISTAT_GOOD
HBA Status                              HOST_NO_ERROR