Graphing failed after changing regular check schedule with new service templates
9 views
Skip to first unread message
Kickstart70
Dec 2, 2010, 2:59:18 PM12/2/10
to GroundWork Monitor CE 6.0
About a month ago, I decided that a number of checks we do should
probably only be checked twice per day, and as such I created a new
service template (called, obviously, "twice-a-day"). I applied that to
these services, and that effort seems to have worked.
However, these services no longer update their graphs. I can't see any
reason for this...the graphs still exist, but have not updated at all.
I've looked through the various logs, but am not sure what I am
looking for, so I haven't found any obvious cause.
Any advice appreciated!
Greg
probably only be checked twice per day, and as such I created a new
service template (called, obviously, "twice-a-day"). I applied that to
these services, and that effort seems to have worked.
However, these services no longer update their graphs. I can't see any
reason for this...the graphs still exist, but have not updated at all.
I've looked through the various logs, but am not sure what I am
looking for, so I haven't found any obvious cause.
Any advice appreciated!
Greg
Dr. Dave
Dec 2, 2010, 5:09:40 PM12/2/10
to GroundWork Monitor CE 6.0
The RRD files that store the graph data have time intervals, often of
just 300 or 600 seconds wide. In addition, there are settings for the
RRD files that say if no update has been received in x seconds then
the time intervals in the RRD must be filled with 'Unknown'. That is
probably what is happening in your case. The RRD files for those
services were created, and the timestamps show they are being updated,
but the RRD engine is saying that the data you are submitting is
spread too far apart. You will need to go into the Performance
Configuration screens and update the RRD create command used for those
services to create appropriate interval widths, and adjust the
'Unknown' setting.
just 300 or 600 seconds wide. In addition, there are settings for the
RRD files that say if no update has been received in x seconds then
the time intervals in the RRD must be filled with 'Unknown'. That is
probably what is happening in your case. The RRD files for those
services were created, and the timestamps show they are being updated,
but the RRD engine is saying that the data you are submitting is
spread too far apart. You will need to go into the Performance
Configuration screens and update the RRD create command used for those
services to create appropriate interval widths, and adjust the
'Unknown' setting.
Kickstart70
Dec 2, 2010, 6:11:19 PM12/2/10
to GroundWork Monitor CE 6.0
Thanks for the reply! (and so happy that you've moved the forum here)
It sounds like you are right...but that does sound like this
necessitates losing all the previous data by a re-creation of the rrd
database?
It sounds like you are right...but that does sound like this
necessitates losing all the previous data by a re-creation of the rrd
database?
Dr. Dave
Dec 2, 2010, 7:14:05 PM12/2/10
to GroundWork Monitor CE 6.0
I confess I'm not 100% sure. To see if the data is still in there,
you could log in to a shell, and then as user nagios run:
rrdtool dump <filename>
This will spit out the data including the averages etc. Scour through
the data to see if you get anything other than NaN or unknown values.
If you do then you will need to read up on how you export RRD data and
then reimport it. There are tools out there to update settings for
RRD files too, so maybe the export and reimport is not necessary.
you could log in to a shell, and then as user nagios run:
rrdtool dump <filename>
This will spit out the data including the averages etc. Scour through
the data to see if you get anything other than NaN or unknown values.
If you do then you will need to read up on how you export RRD data and
then reimport it. There are tools out there to update settings for
RRD files too, so maybe the export and reimport is not necessary.
Reply all
Reply to author
Forward
0 new messages