ML350 G4P Ultra320 SCSI Card I/O Error

[Mike Kulinski]

I am using a Ultra320 Scsi card on a ML350 G4p server. Attached to the two external scsi ports is an HP MSL6000 tape library with 2 960 drives. When I run a backup jobs in BackupExec 9.1 one drive will give me a 0x45d - The request could not be performed because of an I/O device error. I have tried replacing the drive, upgrading the drivers and switching the cables between drives. When I switch cables the problem moves with the cable so it appears to be connected with the card itself. Any help would be appreciated.

[rich pattison]

You need to find out if this is Backup Exec - or a "genuine" hardware failure. Have you tried some other backup utility ?? - (maybe the native Windows backup) - do the L&TT tests run OK ??

Some other questions to ask: - is this a solid fault every time, or intermittant ? - and has it ever worked OK in this configuration? - also, what the O/S & patch level??

"I/O device error" is a kind of "catch-all" description - are there any other messages (Windows event log?? Backup exec logs??)

[Mike Kulinski]

Thanks for your reply. We are running backups on a Windows 2003 Standard with SP1. The error occurs each time I run the backup just at different times in the backup. The Event ID error in Event Viewer is 34113. When I tried to update the firmware of the library and the 2 drives. I was able to do it to everything but the drive that is giving me the error. I tried using NT Backup but it doesn't see the library.

[rich pattison]

From what you say - both drives have run backups successfully, and both have failed - depending on which bus they're attached to. This would indicate you probably don't have a drive/media fault!.

It seems to point to a cable/adapter fault, as you said in your description. The only way you're going to find out for sure is to either replace the adapter, or reconfigure the hardware. One thing you could try is to put both drives on the same bus - this is not really a solution because performance will be poor, but if both drive run OK on one bus, but not on the other then it proves the point - (if you try this - make sure the drives and have different SCSI ID's!!).

A couple of other things to check - both buses should be terminated and RSM (removable storage manager service), should be disabled on the server

Did you upgrade the drives F/W using L&TT ?? - do the drive tests run OK ?? - can you generate support tickets for the drives ??