open-iscsi on SUSE 11

[Stuart Little]

Hi,
Using open-iscsi-2.0.869-8.1 (x86_64) on a SUSE 11.0 based machine,
connecting to an HP D2D.

Generally operation seems fine but I have some issues:-

I cannot get the system to reconnect as start, if I configure open-isci to
start at boot
the system hangs, I have to do a force reboot and boot single user, configure
open-iscsi to off.
Starting open-iscsi by hand always works, but login appears to work on two
interfaces only.

Messages displayed during manual start:-

etc/init.d/open-iscsi start
Starting iSCSI initiator service:
done
Setting up iSCSI targets: Logging in to [iface: default, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aa6.autoloader2.drive,
portal: 192.168.2.85,3260]

Logging in to [iface: iser, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aa6.autoloader2.drive,
portal: 192.168.2.85,3260]
Logging in to [iface: default, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aac.autoloader2.robotic
s, portal: 192.168.2.85,3260]
Logging in to [iface: iser, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aac.autoloader2.robotic
s, portal: 192.168.2.85,3260]
Login to [iface: default, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aa6.autoloader2.drive,
portal: 192.168.2.85,3260]: successful
Login to [iface: default, target:
iqn.1986-03.com.hp:storage.d2dbs.hu191248n0.500110a0002b2aac.autoloader2.robotic
s, portal: 192.168.2.85,3260]: successful

I can change the constraints in the startup to start open-iscsi after
everything else and it will
still hang.
Could the failure to connect on the iser interfaces be a cause of the failure
to start at boot?
If so is there a way to disable the iser interfaces and stick with the tcp
connections?

Another problem was uncovered when a real local tape drive was added, while
status messages suggsted that
the open-iscsi interfaces hand changed from using /dev/st0 to /dev/st1 the
system would not actually
work correctly. I have not tried a reinstall yet, but removal of the offending
local drive let the system
work sensibly again.

Cheers,
Stuart

[Mike Christie]

Yes. Is your system completely hung or does it just not proceed. It
looks like we cannot log into the iser targets. With the default time
outs, it could retry for around 5 minutes, so the system will look hung,
but eventually it would proceed.

Decrease node.conn[0].timeo.login_timeout and
node.session.initial_login_retry_max.

> If so is there a way to disable the iser interfaces and stick with the tcp
> connections?

Set the node.startup to manual for the iser ones.

I think you would want to set things up so the infinniband stuff is
tarted earlier or iscsi is started later. Then you can leave the
node.startup values alone.

>
> Another problem was uncovered when a real local tape drive was added, while
> status messages suggsted that
> the open-iscsi interfaces hand changed from using /dev/st0 to /dev/st1 the
> system would not actually

You lost me. So the tape device is local or accessed through iscsi?

> work correctly. I have not tried a reinstall yet, but removal of the offending
> local drive let the system
> work sensibly again.
>
> Cheers,
> Stuart
>
>
>

> --
>
> You received this message because you are subscribed to the Google Groups "open-iscsi" group.
> To post to this group, send email to open-...@googlegroups.com.
> To unsubscribe from this group, send email to open-iscsi+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/open-iscsi?hl=en.
>
>

[stuart...@ursine.co.uk]

OK, I shall try reducing timeouts to confirm whether the system is
completely hung but probably won't be able to before the New Year.

> >   Another problem was uncovered when a real local tape drive was added, while
> > status messages suggsted that
> >   the open-iscsi interfaces hand changed from using /dev/st0 to /dev/st1 the
> > system would not actually
>
> You lost me. So the tape device is local or accessed through iscsi?

Originally open-iscsi installed to access an HP D2D, so access to it
is via /dev/st0, a psuedo tape device, to actually write data & /dev/
sg3 in this instance to 'control' the autoloader functions of the D2D.
For various reasons they wanted a real local tape drive on the box as
well. When that was installed it became /dev/st0 and the device used
by open-iscsi had chnaged to /dev/st1, however no operations would
work anymore i.e. normally to write data to the D2D you would run

mtx -f /dev/sg3 load 2

This loads a 'slot' on the D2D which is then accessible via /dev/st0
so

tar cvf /dev/st0 <some stuff>

would write data to the D2D.

After installing the new local tape drive the device names reported by
open-iscsi had changed, but the same operations using the new tape
device, /dev/st1, would not work correctly.

[Hannes Reinecke]

Stuart Little wrote:
> Hi,
> Using open-iscsi-2.0.869-8.1 (x86_64) on a SUSE 11.0 based machine,
> connecting to an HP D2D.
>

Is that OpenSUSE 11.0 or SLES11?

I only do _very_ limited testing on OpenSUSE, so SLES is the preferred
platform if you really want to test iSCSI.

I would recommend recompiling the latest open-iscsi code by hand for OpenSUSE.

Sorry for this, but I fear I'm slightly overloaded...

Cheers,

Hannes
--
Dr. Hannes Reinecke zSeries & Storage
ha...@suse.de +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N�rnberg
GF: Markus Rex, HRB 16746 (AG N�rnberg)

[stuart...@ursine.co.uk]

On 13 Jan, 14:31, Hannes Reinecke <h...@suse.de> wrote:
> Stuart Little wrote:
> > Hi,
> >   Using open-iscsi-2.0.869-8.1  (x86_64) on a SUSE 11.0 based machine,
> > connecting to an HP D2D.
>
> Is that OpenSUSE 11.0 or SLES11?

This is SLES11. I still have not had a chance to try the settings
changes to fix the 'work on boot' issue, hopefully will do so soon.

Cheers,
Stuart