Bug#417917: openafs-modules-source: removing openafs kernel module causes kernel error, reinsertion fails

2007-12-18 Thread Russ Allbery
tags 417917 upstream
thanks

Just to update folks on the status of this bug, upstream is hopeful that
this problem has been fixed in the upcoming 1.4.7 release, which should be
going into pre-release relatively soon.

The problem should be less severe with the current packages in unstable,
in that it may be possible to reload the kernel module even when this
problem happens, but there are additional fixes that are in the upcoming
1.4.7 release.

-- 
Russ Allbery ([EMAIL PROTECTED])   http://www.eyrie.org/~eagle/



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#417917: openafs-modules-source: removing openafs kernel module causes kernel error, reinsertion fails

2007-04-12 Thread Russ Allbery
This has now also been reported against the Ubuntu build of 1.4.4, so
apparently while I can't reproduce it, it just happens on some machines
and not on others.  The next step is probably to report it upstream,
although it would still be good if someone who is having this problem
could give me more details.

-- 
Russ Allbery ([EMAIL PROTECTED])   http://www.eyrie.org/~eagle/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Bug#417917: openafs-modules-source: removing openafs kernel module causes kernel error, reinsertion fails

2007-04-05 Thread Stefan Kluth


Subject: openafs-modules-source: removing openafs kernel module causes kernel 
error, reinsertion fails
Package: openafs-modules-source
Version: 1.4.2-6
Severity: grave
Justification: renders package unusable

*** Please type your report below this line ***

After /etc/init.d/openafs-client stop I get from dmesg

...

COLD shutting down of: CB... afs... BkG... CTrunc... AFSDB... RxEvent... 
UnmaskRxkSignals... RxListener...
slab error in kmem_cache_destroy(): cache `afs_inode_cache': Can't free all 
objects
 [c015839f] kmem_cache_destroy+0x7c/0x11e
 [f8dfa5a6] cleanup_module+0x1e/0x53 [openafs]
 [c0135c81] sys_delete_module+0x1ad/0x1d4
 [c014ddc2] remove_vma+0x31/0x36
 [c014e674] do_munmap+0x181/0x19b
 [c0102c11] sysenter_past_esp+0x56/0x79

...

and after /etc/init.d/openafs-client force-start

Found system call table at 0xc02834c0 (pattern scan)
kmem_cache_create: duplicate cache afs_inode_cache
 [c0157fa5] kmem_cache_create+0x41b/0x450
 [c011d97e] printk+0x14/0x18
 [f8debef8] afs_init_inodecache+0x1d/0x2e [openafs]
 [f8debf09] init_once+0x0/0xc [openafs]
 [f8824020] init_module+0x20/0x8f [openafs]
 [c01358c1] sys_init_module+0x16c3/0x1846
 [c0102c11] sysenter_past_esp+0x56/0x79

and openafs said

lapkluth2:~# /etc/init.d/openafs-client force-start
Starting AFS services:FATAL: Error inserting openafs (/lib/modules/2.6.18-4-686/fs/openafs.ko): Cannot allocate 
memory


Failed to load AFS kernel module, not starting AFS


This behaviour started after the recent kernel upgrade to 
2.6.18.dfsg.1-12.  The module was built with m-a auto-install which 
reported no errors.


As a result stop/start of openafs-client is impossible (without a reboot) 
and thus for a mobile user it is unusable.


Cheers, Stefan

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-4-686
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)

Versions of packages openafs-modules-source depends on:
ii  bison   1:2.3.dfsg-4 A parser generator that is compati
ii  debhelper   5.0.42   helper programs for debian/rules
ii  flex2.5.33-11A fast lexical analyzer generator.
ii  module-assistant0.10.8   tool to make module package creati

openafs-modules-source recommends no packages.

-- no debconf information

-- Stefan Kluth, PhD  Wissenschaftler -
-  MPI fuer Physik -  phone:  +49 89 32354 468  -  ATLAS  -
-  Foehringer Ring 6   -  fax:+49 89 32354 305  -  OPAL  -
-- D-80805 Munich, Germany -- e-mail: [EMAIL PROTECTED] -



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#417917: openafs-modules-source: removing openafs kernel module causes kernel error, reinsertion fails

2007-04-05 Thread Russ Allbery
tags 417917 unreproducible
severity 417917 normal
thanks

Stefan Kluth [EMAIL PROTECTED] writes:

 After /etc/init.d/openafs-client stop I get from dmesg

 ...

 COLD shutting down of: CB... afs... BkG... CTrunc... AFSDB... RxEvent... 
 UnmaskRxkSignals... RxListener...

Why are you getting a COLD shutdown instead of a WARM shutdown?  That's
strange.

In any event, this isn't a universal problem.  I can't duplicate this at
all on my system; I can start and stop AFS without any trouble.

 This behaviour started after the recent kernel upgrade to
 2.6.18.dfsg.1-12.  The module was built with m-a auto-install which
 reported no errors.

I'm using the same kernel and the same openafs package that you are.

 As a result stop/start of openafs-client is impossible (without a reboot)
 and thus for a mobile user it is unusable.

Well, I'm not sure why that would make it unusable for a mobile user.  I
can leave my AFS client running on my laptop and it doesn't hurt
anything.  In general, we haven't been treating problems with stopping and
starting the client at that level of severity (it didn't used to work at
all, although a lot of work has been put into it and it's now much
better).

I would like to get it working for you, though.  I'm not sure the best
place to start, however.  One experiment I'd be interested in, if you're
up for it, is to try installing openafs-modules-source from Debian
experimental and seeing if 1.4.4 works any better for you.  Does this
happen every time you try to start and stop the client on your system?

Also, are there any messages before that?  I hardly ever see COLD
shutdowns; I almost always get WARM shutdowns.  I wonder if that may be
somehow related.

-- 
Russ Allbery ([EMAIL PROTECTED])   http://www.eyrie.org/~eagle/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]