hastd: zombies after hooks

[Pawel Jakub Dawidek]

On Sun, Oct 03, 2010 at 03:45:47PM +0300, Mikolaj Golub wrote:
> Hi,
> 
> After recent changes hastd does not garbage collect finished hooks leaving
> zombies and messages in log like below:
> 
> Oct  3 18:21:05 bolek hastd[5144]: Hook is running for 1330 seconds (pid=5223, cmd=[/usr/bin/true connect storage]).
> 
> This is because wait3() in hook_check() is never called (we never call
> hook_check(true)). Below is a patch that works for me. 

We don't call it, because it is not needed anymore. I believe there is a
problem, but I think it must be somewhere else.

In the main process there are two kinds of processes we fork: worker 
processes and hook processes. When SIGCHLD is received we call 
child_exit(), which first check if this is worker process (corresponding
resource is configured). If this is not worker process then it must be
hook process, so it calls hook_check_one().

I'll try to reproduce it and will get back to you.

-- 
Pawel Jakub Dawidek                       http://www.wheelsystems.com
pjd at FreeBSD.org                           http://www.FreeBSD.org
FreeBSD committer                         Am I Evil? Yes, I Am!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-fs/attachments/20101004/dc29ca33/attachment.pgp