[Bug 457184] New: autofs hangs on > 1024 mounts, machine needs reboot -f
https://bugzilla.novell.com/show_bug.cgi?id=457184 Summary: autofs hangs on > 1024 mounts, machine needs reboot -f Product: openSUSE 11.1 Version: RC 1 Platform: Other OS/Version: Other Status: NEW Severity: Normal Priority: P5 - None Component: Network AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: vogt@itwm.fraunhofer.de QAContact: qa@suse.de Found By: --- Hello, I'm using 11.1 rc1 (where can I get rc2 btw?) The following script helps to reproduce the automount bug: #!/bin/sh NET=/net/pxe1/var/tmp for i in Seq 1 2025 ; do echo "/homes/vogt${i} -fstype=bind,ro :${NET}" done This creates a direct mount map with 2025 mounts. eg ./script >/etc/auto.direct and then cat /etc/auto.master: /net /etc/auto.net /- /etc/auto.direct then start automount, I started in on the foreground: /usr/sbin/automount -d -f -t 600 -p /var/run/automount.pid -O ro You should use a real remote host, localhost for example produces another bug, which I'm not (yet) interested in. The go to: /homes/vogt1 check if your mount is really there then trigger the mounts find /homes/ -maxdepth 3 (watch it mount) Then it hangs after sometime. deugging revealed: 1. You have a Zombie process in "mount" root 5111 0.0 0.1 97200 2936 pts/0 Sl+ 13:30 0:01 /usr/sbin/automount -d -f -t 600 -p /var/run/automount.pid -O ro root 9541 0.0 0.0 0 0 pts/0 Z+ 13:31 0:00 [mount] <defunct> 2. automount has 1024FDs open. This looks like a limit, but automount configures itsself to use 10240 fds, which you can check in /proc: pxe1:/proc/5111 # cat limits Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited Max file size unlimited unlimited Max data size unlimited unlimited Max stack size 8388608 unlimited Max core file size unlimited unlimited Max resident set 1758904320 unlimited Max processes 16117 16117 Max open files 10240 10240 Max locked memory 65536 262144 Max address space 2514452480 unlimited Max file locks unlimited unlimited Max pending signals 16117 16117 Max msgqueue size 819200 819200 Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited Heres the lsof from automount, which shows the suspicious 1024 fds. The lsof -p <automount> pxe1:/home/demo1 # lsof -p <automount pid> [...] automount 5111 root 1022r DIR 0,228 0 13029 /homes/vogt<number> automount 5111 root 1023r DIR 0,561 0 13362 /homes/vogt<number> automount 5111 root 1024r FIFO 0,7 0t0 26910 pipe ==> You should be able to reproduce the bug with this script. Btw: rhel52 can mount the directories without hang (but later produces a kernel panic) But ok this bug report is for SuSE of course, but maybe this hint make the debugging a bit easier. I (would) mark this bug critical, but most likely you will set its priority to normal.(Its more critical for me, other users wont hit this problem) regards, Martin -- 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.
https://bugzilla.novell.com/show_bug.cgi?id=457184
User vogt@itwm.fraunhofer.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=457184#c1
--- Comment #1 from Martin Vogt
https://bugzilla.novell.com/show_bug.cgi?id=457184
Andreas Jaeger
https://bugzilla.novell.com/show_bug.cgi?id=457184
User vogt@itwm.fraunhofer.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=457184#c2
--- Comment #2 from Martin Vogt
https://bugzilla.novell.com/show_bug.cgi?id=457184
User mkoenig@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=457184#c3
Matthias Koenig
participants (1)
-
bugzilla_noreply@novell.com