On Thu, 13 Mar 2008, Max Hodgson wrote:

I use Dell 124T tape loaders with Centos. The controlling device on mine is separate to the data device.

Turned out to be at least the need for a new scsi card, which I obtained. I also used a new machine. But your experience below did lead me to review dmesg more carefully, which led to the proper device that happened to not be /dev/st0. mtx did respond positively. It is possible the first "problem" machine could work with the new card, but I have other things to battle :-)

Thanks again.

Scott

e.g.

I use:

$> mtx -f /dev/sg1 status

But for writing to tape:

$> tar cvf /dev/st0 /etc

I think the dmesg told me which device to use to control.

mjh

On 13/03/2008, Scott R. Ehrlich scott@mit.edu wrote:

...

My unit's firmware: library 05.03, tape d22h, shows the device as set to random mode. But mtx -f /dev/st0 status gives an error that google says the device is in sequential mode. dmesg|grep -i hp does reflect CentOS thinks the device is a sequential unit.

I've tried this with Fedora 8, too, and both show the same, so it is either an issue with CentOS/Fedora RPMs, or the version of MTX?

I'm going to try a simple install of Ubuntu next and see if there is any difference.

Is this a bug with the firmware, or with my setup?

I have rebooted the tape drive and machine several times.

I have the drive connected to an Adaptec 29160.

I have another identical drive on another system, with a library firmware of d21h, tape firmware 04.04, and that works perfectly with mtx, though it is connected to a different SCSI card.

Thanks for any help/insights.

Scott _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos