Nagios Support Forum

I have upgraded some of my NCPA agents to 2.1.6 from 1.8.1 and there is a difference in how parameters are passed to plugins when parameters contain '/.'. Please see below:
1.8.1:
[c601018@vhlgnngxi001 ~]$ /usr/local/nagios/libexec/check_ncpa.py -H sclpnekbb001.tvlport.net -t wspwebfarm -M agent/plugin/check_status_cassandra.sh -a "/apps02/tpworkdir/status.log Cassandra_Cluster_Status"
OK: RTDP Cassandra_Cluster_Status is Up

2.1.6:
[c601018@vhlgnngxi001 ~]$ /usr/local/nagios/libexec/check_ncpa.py -H sclppekbb003.tvlport.net -t wspwebfarm -M api/plugins/check_status_cassandra.sh -a "/apps02/tpworkdir/status.log Cassandra_Cluster_Status" WARNING: RTDP apps02 not found

To fix it I had to put the argument containing / in single quotes:
[c601018@vhlgnngxi001 ~]$ /usr/local/nagios/libexec/check_ncpa.py -H sclppekbb003.tvlport.net -t wspwebfarm -M api/plugins/check_status_cassandra.sh -a "'/apps02/tpworkdir/status.log' Cassandra_Cluster_Status"
OK: RTDP Cassandra_Cluster_Status is Up

Unfortunately, I cannot make this change beforehand to my service checks because after putting single quotes around arguments in 1.8.1 does not work:
[c601018@vhlgnngxi001 ~]$ /usr/local/nagios/libexec/check_ncpa.py -H sclpnekbb001.tvlport.net -t wspwebfarm -M agent/plugin/check_status_cassandra.sh -a "'/apps02/tpworkdir/status.log' Cassandra_Cluster_Status"
WARNING: RTDP '/apps02/tpworkdir/status.log' not found

So I have to make the change right after I upgrade the agent. Given the number of agents that are being upgraded at the same time (hundreds) it is a very time consuming task to update service checks as well at the time the change is scheduled.

Hello, @corkyman. We have both versions of NCPA in the lab and we ran our checks successfully with / arguments. We wrapped the whole arguments section in double quotes(like you have in the 1.8.1 example, but also in single quotes and didn't encounter problems either way. I wonder if this is somehow related to the plugin. Can you share the check_status_cassandra.sh with us in the thread?

Attached

Can you also give us an example of the format of the '/apps02/tpworkdir/status.log'?

Attached

I tested various combinations, but unfortunately it is not possible to use the same command for both, the old, and the new agent.

These are two major versions of NCPA, where lots of changes took place. Even if wrapping the arguments in single or double quotes worked the same, the part after the "-M" flag wouldn't work the same...

Since you will be eventually upgrading your NCPA agents to the latest anyway, I would recommend that you set up your commands either as this one: or this one: or: or To avoid doing everything manually, you could use a sed command (find/replace), and modify bunch of configs in bulk. You can then drop them in the /usr/local/nagios/etc/import directory, and run the /usr/local/nagiosxi/scripts/reconfigure_nagios.sh script.

Hope this helps.

Thank you for confirming my findings and a recommendation. I will use a different approach that we've used to modify the -M param--using Nagios API to modify the service checks.
Thanks for your help, you may close this case.