Your message dated Wed, 20 Feb 2008 16:30:01 +0100
with message-id <[EMAIL PROTECTED]>
and subject line Fixed since etch kernel 2.6.16 or so
has caused the Debian Bug report #320181,
regarding smbfs: mount.cifs recurses with automount or subfs
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [EMAIL PROTECTED]
immediately.)
--
320181: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=320181
Debian Bug Tracking System
Contact [EMAIL PROTECTED] with problems
--- Begin Message ---
Package: smbfs
Version: 3.0.14a-2
Severity: grave
Justification: renders package unusable
Symptom: When using automount to mount smb/cifs network shares after a
while (automount expiration time!) the smbd hangs and cannot be killed
(needs reboot). For KDE users this may happen when they right-click on
some file item in Konqueror, but the bug is NOT KDE related. You cannot
get out of this situation (even from a command-line session) without
a reboot!
Analysis: It's simpler to experiment with subfs (but the origin of the
problem was autofs). If you use "mount -t cifs ..." (in a script that
subfs can call on mount request) you get a wonderfull recursion: all
clients that try to access smbd hang (and cannot be killed except by
rebooting). If you kill the "/sbin/mount.cifs" that was started by
"mount -r cifs" another instance pair of mount/mount.cifs will be
created. There is no way out other than reboot. But one can simplify
the thing be calling "/sbin/mount.cifs" directly (inside the subfs
or autofs mount script). In this case the original caller still hangs,
but if you kill the mount.cifs the call fails gently, e.g. you don't
need a reboot.
This looks as if mount.cifs does try to touch the mountpoint before
the samba-mount is complete thereby causing a recursion for autofs
and subfs. Sometimes the first mount works with autofs (but never
with subfs). That mount sometimes works with autofs make the story
a bit confising. But again: after the autofs expiration the problem
becomes more likely to apear even with autofs!
Kernel used: It's the Debian 2.6.11 built from source. No extras.
Subfs taken from SuSE (version 0.92 or so)
-- System Information:
Debian Release: 3.1
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.11-7-amd
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Versions of packages smbfs depends on:
ii libc6 2.3.2.ds1-22 GNU C Library: Shared libraries an
ii libcomerr2 1.37-2sarge1 common error description library
ii libkrb53 1.3.6-2sarge2 MIT Kerberos runtime libraries
ii libldap2 2.1.30-8 OpenLDAP libraries
ii netbase 4.21 Basic TCP/IP networking system
ii samba-common 3.0.14a-6 Samba common files used by both th
-- no debconf information
--- End Message ---
--- Begin Message ---
Package: smbfs
Version: 3.0.24-6etch9
I have no seen any significant cifs/smbfs problems over the last 6 months or
so. Probably thei where caused
Probably the problems where cause by the kernel cifs driver.
Please close the bug, thanks.
--- End Message ---
