Availability Report

[T Verhoef]

I installed nagios on a other system, and that behaved the same. However
....... I changed log-rotation to hourly instead of weekly, and after the
first rotation the avalability reports are there (!)
Put it back to weekly, and it's still OK. so it seems to me it needs a log
rotation first ......


Regards
Tonny


----- Original Message ----- 
From: "Rob Pitman" <rob at pitman.co.za>
To: <nagios-devel at lists.sourceforge.net>
Cc: "Tonny Verhoef" <t_verhoef at hotmail.com>
Sent: Monday, May 12, 2003 1:17 PM
Subject: Re: [Nagios-devel] Availability Report


> Hi,
>
> I had exactly the same problem. In my case, it was because my Nagios
server
> is synchronized to an NTP server, and the clock on my Nagios server is
> gaining a second or two every day. To compensate, NTPD adjusts  the system
> clock backwards in time by a second or two every day.
>
> As I explained in my posting to this mailinglist last Wednesday (May 7),
> there is a bug in "compensate_for_system_time_change" in the file
> nagios.c; it should _not_ adjust the run_time for log rotation events.
This
> bug is present in nagios-1.0 and in nagios-1-x-cvs.
>
> The problem is that the function "compensate_for_system_time_change"
adjusts
> the run-time of the log rotation event (along with all other timed events)
> so that it runs a second BEFORE midnight (I am using daily log rotation).
> Since the log rotation event is a recurring event, another log rotation is
> _immediately_ scheduled for one second later. The second log file is
empty.
> As a result, the "trends", "history" and "notifications" CGIs do not
> display any data for any of the archived logfiles.
>
> Is there someone on this mailinglist who incorporates these bug fixes into
> Nagios? I haven't seen any response to my posting of May 7. I also
> submitted a suggestion about the comment that is associated with scheduled
> downtime (also on May 7).
>
> Regards
> Rob Pitman
>
> On Monday 12 May 2003 09:36, Tonny Verhoef wrote:
> > Hi,
> >
> > I have nagios running, and it does send messages when services and/or
> > systems fail and recover so it seems to works Ok.
> >
> > However, in the availability reports it ALWAYS reports "100% Time
> > Undetermined", even for the desktop PC's I added for test. These go down
> > every evening and are powered-up in te morning. (and messages are send
> > accordingly for those events)
> >
> > Accordingly to the FAQ I have set the settings as follows :
> >   [Nagios.cfg]
> >   log_initial_states=1
> >   retain_state_information=1
> >   state_retention_file=/usr/local/nagios/var/status.sav
> >   retention_update_interval=60
> >   use_retained_program_state=1
> >
> > Why don't I get values in the avalability reports ???
> >
> > Regards,
> > Tonny
>
>


-------------------------------------------------------
Enterprise Linux Forum Conference & Expo, June 4-6, 2003, Santa Clara
The only event dedicated to issues related to Linux enterprise solutions
www.enterpriselinuxforum.com