Nagios Support Forum

I am getting the following errors in our /var/log/messages related to rrdcached:
We are running Nagios XI 5.6.10 on RHEL 7 VMs. Offloaded database with rrdcached 1.4.4

hbouma wrote:I am getting the following errors in our /var/log/messages related to rrdcached:

Code: Select all

Sep 16 11:49:08 XXXXXXXXXXXXXXXXXXXX rrdcached[6547]: queue_thread_main: rrd_update_r (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /Disk_Usage_on__dev_mapper_vg00-var.rrd) failed with status -1. (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /Disk_Usage_on__dev_mapper_vg00-var.rrd: found extra data on update argument: 3.03:4.86)
Sep 16 11:49:09 XXXXXXXXXXXXXXXXXXXX rrdcached[6547]: queue_thread_main: rrd_update_r (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /Disk_Usage_on__dev_mapper_vg00-opt.rrd) failed with status -1. (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /Disk_Usage_on__dev_mapper_vg00-opt.rrd: found extra data on update argument: 0.69:0.95)
Sep 16 11:49:10 XXXXXXXXXXXXXXXXXXXX rrdcached[6547]: queue_thread_main: rrd_update_r (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /_HOST_.rrd) failed with status -1. (/usr/local/nagios/share/perfdata/XXXXXXXXXXXXXXXXXXXX /_HOST_.rrd: expected 4 data source readings (got 1) from 1600270238)

We are running Nagios XI 5.6.10 on RHEL 7 VMs. Offloaded database with rrdcached 1.4.4

Did you change the command for these to return a different quantity of performance data that was there before? It seems you are getting a different amount of perfdata that the RRDs expect

This is being pulled from the NCPA agent checks using NCPA version 2.1.5. It is using the built in disk check functionality.

hbouma wrote:This is being pulled from the NCPA agent checks using NCPA version 2.1.5. It is using the built in disk check functionality.

Can you share the command it is using?

Also, could you show a screenshot of the advanced tab for the service as well as a pic of the performance graph for the service

Thanks

Here is an example of one of the commands:

check_ncpa.py -H HOSTNAME -t 'TOKEN' -P 5693 -M 'disk/logical/|var|log' -w 80 -c 90

scottwilkerson wrote:as well as a pic of the performance graph for the service

thanks

Performance graph is blank.

Ok, so this performace graph just has one metric "used" and the current check command you are using contains 3 metrics used, free, total

At some point the command must have changed.

The only way to rectify this is to remove the rrd for this service from and let it get re-created with all the metrics

So, this has happened across multiple checks for multiple servers. Is it possible that something messed up the metric info on so many?

hbouma wrote:So, this has happened across multiple checks for multiple servers. Is it possible that something messed up the metric info on so many?

That does seem odd, have you made any changes across them? or commands?