Bug (proposed) - lockd_down: lockd failed to exit
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Below the contents of a bug report I plan to submit against sysvinit.
The bug concerns an error message on console during poweroff when unmounting
NFS-shares.
Any comments/suggestions welcome.
Frans Pop
From: Frans Pop <aragorn@tiscali.nl>
To: submit@bugs.debian.org
Subject: umountnfs.sh: lockd_down: lockd failed to exit
Cc: nfs-common@packages.debian.org
Package: sysvinit
Version: 2.84-2woody1
This bug also concerns package nfs-common, version 1:1.0-2woody1.
During poweroff I always get the following message on console:
unmounting remote filesystems... \
lockd_down: lockd failed to exit, clearing pid \
done
This is on a desktop client for a Debian Woody NFS-server.
I have NFS-kernel support compiled in the kernel, so when the system is up I
have (from ps alx):
F UID PID ... WCHAN STAT ... COMMAND
140 1 129 poll S /sbin/portmap
040 0 135 rpciod SW [rpciod]
040 0 136 svc_re SW [lockd]
140 0 214 select S /sbin/rpc.statd
It seems to me the problem could be the order in which scripts from
/etc/init.d are run.
K81nfs-common - stops /sbin/rpc.statd
S10portmap - stops /sbin/portmap
S19devfsd
S20sendsigs
S31umountnfs.sh - runs umount -tnfs -f -a -r; causing the error message
By the time umountnfs.sh is run, only [rpciod] and [lockd] are still running.
Part of the problem could be that, because I have NFS-kernel support,
rpc.lockd is not run on my system and umount thinks it should be running. I
think this may be part of the problem because it looks like umountnfs.sh
tries to remove a PID (for rpc.lockd?).
I am running /devfs. This could also be part of the problem as it is stopped
(S19devfsd) before umountnfs.sh is run.
System information
==================
Debian Woody with Linux 2.4.20 (i686)
libc6 Version: 2.2.5-11.5
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQE/VeTxgm/Kwh6ICoQRAoApAJwMnwedR1ywG/SOnONmWw5QXY9+twCgn3Qw
q8efkdNAg4zrcbVzHZM9XW8=
=D6+T
-----END PGP SIGNATURE-----
Reply to: