nsrjb takes very long time

[litauer]

Since upgrading our (linux) networker server from kernel 2.4 (SuSE Linux 9.0) to kernel 2.6 (SuSE Linux 9.3), nsrjb needs a lot of time for nearly any action, e.g.

holmes:~ # nsrjb -v -n -l LTO-0001
setting verbosity level to `1'
------------- about 4 minutes waiting
nsrjb: Looking at device `/dev/nst0' id 0.
device = /dev/nst0, id = 0, reserved = 1, busy = 0, mode = 0
Jukebox handle being opened at port scsidev@1.6.0
box_inventory:
Jukebox handle being closed at port scsidev@1.6.0
box_inventory_free:
Jukebox handle being opened at port scsidev@1.6.0
Slot `1' is being loaded into drive `/dev/nst0'
box_display: `LTO-0001' for drive `/dev/nst0'
Jukebox handle being closed at port scsidev@1.6.0
load sleep for 5 seconds

My syslog contains some messages:
Apr 28 10:38:16 holmes kernel: program nsrmmd is using a deprecated SCSI ioctl, please convert it to SG_IO

Is this the reason for the timeouts? What can I do against it?

 

[steff13]

Hello,

I have the same problem, what can I do?

Thanks, Steff

 

[NWNinja]

Seems like a lot of applications are having problems with deprecated SCSI ioctls with the 2.6 kernel. What NetWorker version are you running? If you are not running 7.2.1 you could always try and upgrade, but i have not seen anything about this problem on Legato´s site so i suspect it won´t work any better. And as far as i can tell from the compability guide, SuSE 9 does not seem to be supported as a NetWorker server.

Cheers!
Maverick