Red Hat Bugzilla – Bug 427117
idle direct mounts unmounted by HUP to automount
Last modified: 2008-01-04 00:45:38 EST
Description of problem:
sending a HUP signal to the automount process unmounts all the idle direct mounts.
Version-Release number of selected component (if applicable):
kernel 2.6.18-53.1.4.el5 and autofs-5.0.1-0.rc2.55
Steps to Reproduce:
1. start automount with a large (6500 entry) direct map & debug logging
2. send a HUP signal to the automount pid
3. watch syslog, observe falling count of direct map entries
direct mounts are removed
daemon rereads config files, picks up any new entries
I've been discussing this with Ian Kent already - he believes it's already fixed
in the version in CVS.
(In reply to comment #0)
> I've been discussing this with Ian Kent already - he believes it's already fixed
> in the version in CVS.
I've uploaded the current i386 and x86_64 CVS builds to
If you need different archs let me know.
As always we must keep in mind that these are the current CVS
development and haven't completed the QA process so there may
be unexpected problems. But then that's what testing is about.
Your help is appreciated.
(In reply to comment #1)
> I've uploaded the current i386 and x86_64 CVS builds to
> If you need different archs let me know.
That should read www.kernel.org not www.eu.kernel.org.
Created attachment 290809 [details]
automount debug output from autofs-5.0.1-0.rc2.79
I've downloaded autofs-5.0.1-0.rc2.79 and installed it on a test host. I'm still
seeing the same problem, but the debug output is a lot more verbose now. I'm
attaching the log to this bug.
(In reply to comment #4)
> I've downloaded autofs-5.0.1-0.rc2.79 and installed it on a test host. I'm still
> seeing the same problem, but the debug output is a lot more verbose now. I'm
> attaching the log to this bug.
Thanks for the log.
Looks like I've been able to duplicate the problem.
I'll dig a bit deeper.
*Sigh*, the included map stuff is a bit tricky.
I'm going to mark this bug as a duplicate of bug 336961 because
the symptoms are the same and it looks like a case that was
missed in the correction for that bug.
*** This bug has been marked as a duplicate of 336961 ***