On 07/31/2010 10:52 PM, sf...@users.sourceforge.net wrote:
>> I am afraid you won't be happy even if I develop a new feature such like
>> - ioctl interface to get the opened file path
>> - /sys/fs/aufs/*/opened_files
>> - or something.
>> Because it is essentially equivalent to Magic SysRq.
>>
> I've got a tiny idea.
> Try commenting-out a line in fs/aufs/branch.c:au_br_del().
>
> br = au_sbr(sb, bindex);
> i = atomic_read(&br->br_count);
> if (unlikely(i)) {
> AuVerbose(verbose, "%d file(s) opened\n", i);
> --> //goto out;
> }
>
> And specify "verbose" when you delete a branch, then you may be able to
> see what you want.
>
>
> J. R. Okajima
>
Sorry for the delay.
I did give this a shot. It's definitely closer but still doesn't
pinpoint the blocking process for all scenarios.
If it output the process id(s) which caused it to fail it would be
perfect and it wouldn't require aufs to be compiled with debug to parse
the the verbose sysreq output.
This is not a high priority though since killing all processes accessing
the union is manageable.
Lou Gosselin
------------------------------------------------------------------------------
This SF.net email is sponsored by
Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev
