Hi
We are providing a VPN-based NMS server for our customer that is monitoring their DC equipment. In particular they are interested in the relationship between processor usage and bandwidth utilisation. Using the graph explorer we are able to co-relate this data but the resulting graph is 'skewed' by 5 minutes. I would imagine this is due to the fact that the proc usage is a direct snmp get by Nagios XI and the bandwidth usage is an snmp get by mrtg then written to a file which is then read by Nagios XI thus resulting in a 5 min skew.
Obviously the customer is not interested in how the NMS is gathering the stats but will require that the graphs line up else the data is bogus i.e. that the mrtg stats are 5 mins behind.
Is there a way of syncing these upo so that the mrtg information is correct. Perhaps the timings on mrtg can be used?
See attached.
Many thanks
Traffic and CPU data skewed by 5 mins
Traffic and CPU data skewed by 5 mins
You do not have the required permissions to view the files attached to this post.
Re: Traffic and CPU data skewed by 5 mins
The problem is that the checks don't occur at the same time, the MRTG RRD is queried by the service check and put into a new RRD, so the time points are aligned with when the service check ran.
The only way (currently) that you could get that to sync up would be to have one service do both of the checks and submit the results at the same time, so essentially write a wrapper script that would call both but submit separate results at the same time:
Running two of these: http://old.nagios.org/developerinfo/ext ... and_id=114
The only way (currently) that you could get that to sync up would be to have one service do both of the checks and submit the results at the same time, so essentially write a wrapper script that would call both but submit separate results at the same time:
Running two of these: http://old.nagios.org/developerinfo/ext ... and_id=114