Exabyte VXA-172 drive no longer seen by BE or Netware

[tjubb]

Out of the blue the other day my VXA-172 drive stopped being recognized both by Backup Exec 9.2.1401.5 and NetWare 5.1 SP8.  The BE job log showed this error:
"Write error occurred on media - 24615".

Exatool says there is "no supported SCSI devices found"

"List Storage Adapters" does not list the Exabyte drive anymore.

At the moment I am not in a position to reboot the server and see if the drive re-aquaints itself with NetWare.

Is there anything else I can do to get this drive recognized again?

Thanks,
Tom

[criley]

You could try the following commands:

SCAN FOR NEW DEVICES
SCAN ALL LUNS

Then do another LIST STORAGE ADAPTERS to see if NetWare picks it up.

[tjubb]

I just now tried all those console commands and the drive is still not detected.  I am going to reboot the server tonight and see if it re-appears.  If that doesn't work then I guess I have a drive problem and will have to call Exabyte.

[R_H_2]

It may sound strange, but try cleaning the drive with a cleaning cartridge that you know is good then try those commands again before you reboot.

Message Edited by R H on 06-28-200711:45 AM

[tjubb]

The cleaning cartridge trick did not work but the server reboot did.  I guess it was something in NetWare's drivers or scsi board that was the problem.

Once I got the drive recognized I updated the firmware to the VXA-172 and it timed out.  The Exatool said to restart the drive in 5 minutes.  I did that and the drive was again undetected.  So I did the "scan all luns" and "scan for new devices" and the drive was detected again.  Even though the firmware update timed out, the Exatool utility reported the new firmware version.  Strange.

Hopefully tonight's backup job works properly as it's been a week since the last good backup.  I did a test backup/restore and that was successful.

Thanks for the help!
Tom

[R_H_2]

After a firmware update, the device usually resets itself.  Sometimes software / OS can handle it.  Sometimes it can't and the app trying to access it hangs or times out.

How did the backup go?

[tjubb]

well, I thought the problems were behind me.  After doing a small test backup successfully last week I felt pretty good about things.  I came into the office this morning and the Friday night backup failed:
NDMP_NO_DEVICE_ERR (16) Unable to open target media device

I've seen this error before but don't recall what I did to fix it.  I'll be scanning through the knowledge base for clues.

[R_H_2]

It sounds like you may have some SCSI errors happening.  REALLY tough to diagnose on a Netware box as there are no useful logs (that I am aware of) that you can look at.

Is this a stand alone drive or is it in an autoloader/library?

[tjubb]

This is a stand alone drive.
I've read that sometimes this error can happen with a bad tape.  My Friday tapes have caused problems in the past but have been sporadic.  I might see what tonight's backup brings with a different tape.  Would be a nice easy fix to just replace a few tapes.

[R_H_2]

True with the bad tape thing.  Definately worth trying a few jobs on new tapes.

Let us know how it goes.

[tjubb]

success!  I finally had a good backup job complete last night.
I did a bestop/bestart, changed the scsi id on the drive (not sure if that had any impact), power cycled the drive and BE recognized the drive and did a successful backup/restore, even on the tape I thought was bad.

Hope things stay this way for awhile.  This has been the bane of my existance over the past few weeks!

Thanks all!
Tom

[tjubb]

ughhhh!  The dreaded NDMP_NO_DEVICE_ERR is back again.  The problem began the other night when the backup ended prematurely.  It only grabbed about half of the  data it should have before timing out:
"A timeout has occurred on drive HA:1 ID:8 LUN:0 Exabyte VXA-172."  The error also referenced a scsi bus timeout which went on to mention that it could be caused by a drive that needs cleaning, incorrect termination, or faulty device.

Then the next night the job failed altogether with the NDMP_NO_DEVICE_ERR error.

I tried a scan for new devices, scan all luns and list storage adapters.  A couple observations:  When doing a scan for new devices, the server was working so hard that users reported disconnects and slow network response.  It eventually freed up resources and users were able to work normally again.  After tdoing a list storage adapters, the exabyte drive showed up about 6 times.  I don't ever remember seeing multiple instances.

I have purchased brand new tapes and will try my luck with those as I have cleaned the drive 6 time this month already.  If the new tapes produce the same problem I guess it's a hardware issue, either drive or scsi board, I'm not sure.  I'm also going to try another server reboot but I'd rather not make rebooting a regular practice.

Not sure why it seemed to be working for 2 weeks and then crap out again.

T

Message Edited by tjubb on 07-20-200708:45 AM

[R_H_2]

Seeing a device multiple times usually means that the device is shading a SCSI ID with something else.  Double check that it isn't set to ID 7.

Is there anything else on this SCSI bus?

[tjubb]

yes, the scsi id was set to 7.  I changed it to 8.  Is there a conflict for keeping it on 7?
I don't believe there is anything else on the bus.  Will a "list storage adapters" show what's on each bus?

[tjubb]

ok, I think I have this working again.  Through various "scan" commands, bestop and bestarting, cleaning out the multiple drive listing in BE, bestop and bestarting again, and using a fresh brand new tape, I was able to backup a few hundred megs without error.  I'll see what the full backup tonight does.  Hopefully I don't see this problem for a long time.  It's been a huge annoyance for me as of late.

[R_H_2]

The SCSI controller should always be set to ID 7.  ID 8 should due.

[tjubb]

the nightmare continues!

While I can get Netware and Backup Exec to recognize the drive.  I have been getting all sorts of errors.  I reluctantly blew away the BE directory so I don't have the log to list the error codes.  I do recall lots of media errors even on brand new or very young tapes.   But my main problem now is the locking up of the server when running Backup Exec.  When I create a test job and it starts, it seems to hang on the preliminary step of the backup.  Something about searching the tape or blocks.  At that point, nobody is able to access the network drives, GroupWise, etc.  Even when I run the Exatool for diagnostics on the drive, it too brings the server to it's knees and locks everyone out.  All my headaches have seemed to begin with the last 2 firmware updates.  Out of the box the drive was fine.  Is there any way to go back to a previous firmware?  I know this isn't an Exabyte forum.  They don't have a forum.

So, I'm really having difficulty pinpointing this as a BE issue or an Exabyte issue.  Since the Exatool program has fits too I think it may be hardware.  As it stands it is very hard to troubleshoot since it's so unstable and locking people out of the network.

Would love any kind of insight or tips before I lose my mind.

Tom


Update:
Ok, just found the most recent error:
Unknown Media Management Error 24683.
Not much is found in the knowledge base on this one.


Message Edited by tjubb on 09-27-2007 01:38 PM

Message Edited by tjubb on 09-27-2007 01:38 PM