Disk Enclosures
1748227 Members
4366 Online
108759 Solutions
New Discussion юеВ

Re: Disconnecting Autoraid 12H

 
Rainer von Bongartz
Honored Contributor

Disconnecting Autoraid 12H

I'm trying to disconnect a 12H from my K360.
All data has been migrated to a SAN array, LUN's on 12 H have been deleted.

Now I want to disconnect and power down the 12H.
I shut down the K360, disconnected the 12H from the HSC controller, terminated the SCSI adapters and re-booted the box.


I got a crash (TOC) while loading the kernel. No dumps could be written, because IO and dump device was 'not configured at the time of crash'.

After re-connecting the 12 H the K360 booted up.


What am I missing in the procedure to disconnect the array ??

Regards
Rainer

He's a real UNIX Man, sitting in his UNIX LAN making all his UNIX plans for nobody ...
4 REPLIES 4
Peter Godron
Honored Contributor

Re: Disconnecting Autoraid 12H

Rainer,
what is your crash dump device configured as?
/sbin/crashconf and /etc/rc.config.d
Regards
Rainer von Bongartz
Honored Contributor

Re: Disconnecting Autoraid 12H

Peter

crash dump is enabled

# CRASHCONF_ENABLED: Configure crash dumps at boot time.
CRASHCONF_ENABLED=1

and dump device is:

DEVICE OFFSET(kB) SIZE (kB) LOGICAL VOL. NAME
------------ ---------- ---------- ------------ -------------------------
31:0x025000 52064 1048576 64:0x000002 /dev/vg00/lvol2

Regards
Rainer
He's a real UNIX Man, sitting in his UNIX LAN making all his UNIX plans for nobody ...
Peter Godron
Honored Contributor

Re: Disconnecting Autoraid 12H

Rainer,
I assume you stopped the array and removed from /etc/fstab.
To me it seems on startup the K360 is still looking for the array. The way forward would be to first remove all traces of the array then get the crash dump to work and analyse the dump file.
Steve Lewis
Honored Contributor

Re: Disconnecting Autoraid 12H

Are you sure you didn't have a 2nd swap area on the 12H?
Even though LUNs have been deleted I have heard of some arrays responding to 'are you there?' probes via the old device files.
Boot in single user mode and edit /etc/fstab.
Then check that someone didn't previously add any 12H LUNs to vg00. That would be very bad and probably a case for maintenance mode.