[Nagiosplug-devel] FW: Check_disk plugin update

Thomas Guyot-Sionnest dermoth at aei.ca
Sat Nov 15 00:40:08 CET 2008

Hash: SHA1

On 13/11/08 10:14 PM, Thomas Guyot-Sionnest wrote:
> On 13/11/08 09:50 PM, Thomas Guyot-Sionnest wrote:
>> On 13/11/08 01:52 PM, Erik Welch wrote:
>>> How did your testing go? Is there more work/testing that you need from me?
>>> We have a rather large and complex automount environment, so we tend to hit some interesting corner cases on just about every tool we use...
>> Sorry; I've been busy with work and converting the repository for a
>> SVN -> Git migration coming soon.
>> The last patch works, although it obviously leak memory for each -p
>> invocation... I tried freeing the struct first but for reasons that I
>> still don't understand it fails horribly... So I finally got it working
>> by using a temporary struct for that code block.
>> I'd still like some feedback from other developers, as I don't like much
>> my current code, and I would especially like to know why it failed when
>> freeing the  global array. I could also make it a configure option,
>> because I doubt many people will use that anyway.
>> Oh, out of curiosity, since the stat call mounts the partition,
>> shouldn't it be mounted on further checks? What is your check interval
>> and max_check_attempt?
> Oops, I totally forgot to post the url to my Git repo with this patch:
> http://solaris.beaubien.net/cgi-bin/gitweb.cgi/nagiosplug-dermoth.git
> It's not in trunk yet, and I may amend it.

Nevermind - this patch has no effect because I forgot to use temp_me
instead of mount_list in np_set_best_match. It should work with this
little change:

in the case 'p':, change:

np_set_best_match(se, mount_list, exact_match);


np_set_best_match(se, temp_me, exact_match);

I'm still working on it though, mount_list should be freeable :)

- --
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org


More information about the Devel mailing list