Memory Usage graph stopped plotting even though check is OK

[solarmon]

Hi,

Nagios XI version: 5.6.6

I had converted a "Memory Usage" service check using "check_ncpa.py" for a specific host in to a template "Memory Usage" service check using "check_snmp_storage_wizard.pl", which is mapped to a Host Group that has the host in it.

Since then, the 'Memory Usage' graph for the host has stopped plotting. The service checks returns OK, but the data is not being plotted.

I believe this is because the 'Memory Usage' graph for the same host is called the same but the 'format' of the graph is different - i.e was initially created in the the format used by "check_ncpa.py", but now "check_snmp_storage_wizard.pl" is expecting the graph to be in its format?

How do I 'reset' or delete a particular service graph for a particular hosts, assuming this is the way to resolve the issue?

[scottwilkerson]

You would need to delete the .rrd file for this check, it would be located in a path like

Code: Select all

/usr/local/nagios/share/perfdata/<HOSTNAME>/Memory_Usage.rrd

Then the .rrd file will recreate with the appropriate amount of datasources in about 5-15 minutes. This will remove any historical information that was in the RRD file befor the change but is the only way when changing plugins

[solarmon]

Thank you! Deleting the RRD file resolved the issue.

[solarmon]

I assume any files in these directories that have stale timestamps should be removed?

I assume Nagios XI do this housekeeping automatically and for you?

[scottwilkerson]

I assume any files in these directories that have stale timestamps should be removed?

You can, if you are no longer using the checks

I assume Nagios XI do this housekeeping automatically and for you?

No it does not, primarily because sometime people want to turn off a host/service for a while and then turn it back on and would like historical performance data.