RE: [Perfparse-users] nagios restarting

[Yves M. <yme...@li...>]

> The problem I see restarting nagios every 10 minutes (in my case with p=
p),
> is that nagios log grows too much, since I get around 700 services and =
hosts
> alerts saying "OK". (I monitor 100 hosts and 600 services).

Garry said that with a lot of hosts and services, you sometimes need to f=
lush the logs
(and rotate them or compress them or anything...).
I agree with him.
I'm working on something that will allow to send the data without the nee=
d to restart
nagios every 10 minutes. But I suggest that you do it every 24h anyways. =
It can also fix
nagios problems when you compiled the embedded perl interpreter in nagios=
, and refresh
the state of the services... but this is beyond our subject :)

Here is some suggestion before we release the new version with such a fea=
ture (not soon):
1/ Use perfparse every 10 minutes without --delete* options. This will se=
nd the perf data.
2/ Use perfparse every 24 hours *with* the --delete option. This will emp=
ty the perf log
(and send the perf data).

I'm interested in how many lines you get in such a file in 24h, the size =
of the file,
and the size when you compress it with gzip -9.
I can see what priorities to set on my developments :)

Yves (main contributor on perfparse)


--=20
- Homepage    - http://ymettier.free.fr - http://www.logicacmg.com -
- GPG key     - http://ymettier.free.fr/gpg.txt                    -
- Maitretarot - http://www.nongnu.org/maitretarot/                 -
- GTKtalog    - http://www.nongnu.org/gtktalog/                    -