jpff
July 28, 2013, 2:40pm
#1
I have three machines running 12.2 in 64bit format. One of them has problems when booting. About 50-60% of the times it crashes with a backtrace relating to nfs; see Jul 28 11:57:57 tippett syslog-ng[696]: syslog-ng starting up; version=‘3.3.5’ - Pastebin.com
first crash is
Jul 28 11:58:01 tippett kernel: 27.853983] FS-Cache: Netfs ‘nfs’ registered for caching
Jul 28 11:58:01 tippett sm-notify[1516]: Version 1.2.6 starting
Jul 28 11:58:01 tippett kernel: 27.869866] BUG: unable to handle kernel NULL pointer dereference at (null)
and part of backtrace
Jul 28 11:58:01 tippett kernel: 27.953480] Call Trace:
Jul 28 11:58:01 tippett kernel: 27.956566] <ffffffff81593e85>] notifier_call_chain+0x45/0x60
Jul 28 11:58:01 tippett kernel: 27.959623] <ffffffff81066d05>] __blocking_notifier_call_chain+0x55/0x80
Jul 28 11:58:01 tippett kernel: 27.962655] <ffffffffa0cf17d9>] rpc_fill_super+0xe9/0x1a0 [sunrpc]
Jul 28 11:58:01 tippett kernel: 27.965689] <ffffffff81161399>] mount_ns+0xa9/0xe0
Jul 28 11:58:01 tippett kernel: 27.968686] <ffffffff81161ae5>] mount_fs+0x45/0x1d0
The machine mounts one nfs disk on boot.
Is there a fix or a work-around?
jpff:
I have three machines running 12.2 in 64bit format. One of them has problems when booting. About 50-60% of the times it crashes with a backtrace relating to nfs; see Jul 28 11:57:57 tippett syslog-ng[696]: syslog-ng starting up; version=‘3.3.5’ - Pastebin.com
first crash is
Jul 28 11:58:01 tippett kernel: 27.853983] FS-Cache: Netfs ‘nfs’ registered for caching
Jul 28 11:58:01 tippett sm-notify[1516]: Version 1.2.6 starting
Jul 28 11:58:01 tippett kernel: 27.869866] BUG: unable to handle kernel NULL pointer dereference at (null)
and part of backtrace
Jul 28 11:58:01 tippett kernel: 27.953480] Call Trace:
Jul 28 11:58:01 tippett kernel: 27.956566] <ffffffff81593e85>] notifier_call_chain+0x45/0x60
Jul 28 11:58:01 tippett kernel: 27.959623] <ffffffff81066d05>] __blocking_notifier_call_chain+0x55/0x80
Jul 28 11:58:01 tippett kernel: 27.962655] <ffffffffa0cf17d9>] rpc_fill_super+0xe9/0x1a0 [sunrpc]
Jul 28 11:58:01 tippett kernel: 27.965689] <ffffffff81161399>] mount_ns+0xa9/0xe0
Jul 28 11:58:01 tippett kernel: 27.968686] <ffffffff81161ae5>] mount_fs+0x45/0x1d0
The machine mounts one nfs disk on boot.
Is there a fix or a work-around?
Did this problem just start, perhaps after an update? The following terminal command can find the most recent updates:
rpm -q --all --last
Its possible a kernel update might help, since you are down at kernel 3.4, have a look here for help:
openSUSE and Installing New Linux Kernel Versions - Blogs - openSUSE Forums
And for anything to do with Zypper and app updates, have a look here:
Zypper Command - Zypper Package Management Menu System - Version 2.00 - Blogs - openSUSE Forums
Thank You,
jpff
July 28, 2013, 6:53pm
#3
I update this machine regularly, usin zypper; usually weekly. I think it started after
nfs-kernel-server-1.2.6-2.16.1 Sat 15 Jun 2013 23:14:48 BST
but I cannot be totally sure when it started.
I thought it was running the most uptodate 12.2. My experience of 12.3 is fairly uniformly negative – only just got my 12.3 machine to actually boot without lots of hand holding (especially over nfs and X), and there have been other issues locally.
I did wonder if not mounting the remote disk on boot but doing it a little later might work. It has all the hallmarks of a timing issue.
I will look at the kernel link
jpff:
I update this machine regularly, usin zypper; usually weekly. I think it started after
nfs-kernel-server-1.2.6-2.16.1 Sat 15 Jun 2013 23:14:48 BST
but I cannot be totally sure when it started.
I thought it was running the most uptodate 12.2. My experience of 12.3 is fairly uniformly negative – only just got my 12.3 machine to actually boot without lots of hand holding (especially over nfs and X), and there have been other issues locally.
I did wonder if not mounting the remote disk on boot but doing it a little later might work. It has all the hallmarks of a timing issue.
I will look at the kernel link
I have a bash script on loading Samba you can find at the link below. The interesting part can be found in the comments with a couple of actions that might help with NFS as well.
S.A.C.T. - Samba Automated Configuration Tool - Version 1.06 - Blogs - openSUSE Forums
One was on nmb failing on boot as the same fix might work here and the other is when is the interface activated. At boot or when the Desktop loads? The latter might happen with Wireless I have found. Last, moving activation of NFS to later might be helpful as well. You can put a bash script in /usr/local/bin for instance you could run manually.
Thank You,