http://bugzilla.novell.com/show_bug.cgi?id=605195 http://bugzilla.novell.com/show_bug.cgi?id=605195#c Jason Douglas changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jdouglas@novell.com, | |pmillett@novell.com AssignedTo|jdouglas@novell.com |jfehlig@novell.com -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=605195 http://bugzilla.novell.com/show_bug.cgi?id=605195#c2 Jan Beulich changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P3 - Medium Status|NEW |NEEDINFO Found By|--- |Community User InfoProvider| |koenig@linux.de --- Comment #2 from Jan Beulich 2010-05-17 09:10:34 UTC --- It is not clear under what conditions this can happen; simply adding a NULL pointer check to the CDROM_GET_CAPABILITY case of blkif_ioctl() de-referencing gd without understanding how that state can be reached seems inappropriate to me. Therefore it would be necessary to know what precisely your system is doing when that issue arises, what kind of block device (disk vs. CD-ROM) it is that is getting attached to Dom0, and whether this was a one-time observation or is repeatable (all of this ideally so that we can reproduce it - your configuration seems to be far from trivial). -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=605195 http://bugzilla.novell.com/show_bug.cgi?id=605195#c4 Harald Koenig changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW InfoProvider|koenig@linux.de | --- Comment #4 from Harald Koenig 2010-05-17 17:54:05 UTC --- (In reply to comment #2)

Therefore it would be necessary to know what precisely your system is doing when that issue arises, what kind of block device (disk vs. CD-ROM) it is that is getting attached to Dom0, and whether this was a one-time observation or is repeatable (all of this ideally so that we can reproduce it - your configuration seems to be far from trivial).

for now I've looked into the other log files and here is some more information about that incident from last week's test, attached as text file because of that broken web text input stuff :-( the most important details from the attachment: May 12 16:10:49 os4 kernel: [ 4544.252519] BUG: unable to handle kernel NULL pointer dereference at 00000000000002d0 May 12 16:10:49 os4 udevd-work[13239]: 'cdrom_id --export /dev/xvdp' unexpected exit with status 0x0009 May 12 16:10:49 os4 kernel: [ 4544.252686] IP: [<ffffffffa0117d5f>] blkif_ioctl+0x2f/0xe0 [xenblk] where /dev/xvdp is the block device for the root disk for the domU via iscsi: disk=[ 'iscsi:iqn.2010-04.de.science-computing:os-debian40-flat.vmdk,hda,w' ] I'll have some time slots for some more testing within the next few days (hopefully with 11.3 in dom0 too), so I'm about suggestions how to get the relevant data... so far I guess it might be interesting (see attachment with more log file info):... a) manual iscsi login to the root disk and run "cdrom_id --export /dev/whatever" on the "regular" scsi device will the the strace output of this cdrom_id run be any helpful ? b) if cdrom_id works (no Nullpointer) for "normal" iscsi /dev/sdX device, is there an easy way to manually to the setup steps for /dev/xvdp "the XEN way" to compare results here ? c) one test will be to start up all the domUs manually one my one to separate any trouble events and logfile entries... d) I do have a local copy of that os-debian40-flat.vmdk disk image, so I could compare iscsi vs. local disk if necessary [ FYI: right now there are 10 domU with local disk images and 2 domU with iscsi disks -- all will be started automatically at dom0 boot time... ] more ideas or info request before I start my next test slot ? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=605195 http://bugzilla.novell.com/show_bug.cgi?id=605195#c6 Jan Beulich changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO InfoProvider| |koenig@linux.de --- Comment #6 from Jan Beulich 2010-05-19 12:45:04 UTC --- Those "cdrom" kernel messages are harmless (and occur always). Restoring needinfo, for you to provide information on whether and under what circumstances this occurs. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=605195 https://bugzilla.novell.com/show_bug.cgi?id=605195#c7 Jan Beulich changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |CLOSED InfoProvider|koenig@linux.de | Resolution| |NORESPONSE --- Comment #7 from Jan Beulich 2010-09-28 15:22:41 UTC --- The requested information has not been provided for more than 4 weeks. The bug gets therefore closed as NORESPONSE. Please reopen the bug if you can supply the requested information. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.