Page 1 of 1
Many NRPE client checks do not work on REDHAT Linux
Posted: Thu Feb 25, 2016 1:19 pm
by dlukinski
This is a continuation of the last year topic about NRPE (one of mine)
Given: stock NAGIOS XI 5.2.3 appliance + stock Linux NRPE agent from Nagios assets
Upon install of the agent with or w/o pre-reqs (like those required for SuSe) results are always the same for multiple servers we had attempted.
Please review JPG attached (checks that do not work) and CFG files from the same host after NRPE agent install
Please help with finding best way to fix this and automate NRPE process more or less (by supplying updated CFG files after install?)
Thank you
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Thu Feb 25, 2016 1:50 pm
by rkennedy
Can you try to execute one of the failing commands over the CLI from the Nagios machine, and let us know the result? This should give us a bit more information as to what's going on.
On the client machine, please run a tail -n 500 /var/log/messages and post the output?
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Mon Feb 29, 2016 8:49 pm
by rajasegar
Looks familiar.
You might be missing the glibc.x86 library.
Run the plugin manually from the server to see the error message.
Then ldd the plugin (if applicable) to see if there is any missing dependencies.
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Tue Mar 01, 2016 9:07 am
by dlukinski
rajasegar wrote:Looks familiar.
You might be missing the glibc.x86 library.
Run the plugin manually from the server to see the error message.
Then ldd the plugin (if applicable) to see if there is any missing dependencies.
Could you help with listing what has to run (we have no access to these servers, everything done by the vendor)
- need some level of detail in the instruction to forward
Please
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Tue Mar 01, 2016 11:45 am
by rkennedy
rkennedy wrote:Can you try to execute one of the failing commands over the CLI from the Nagios machine, and let us know the result? This should give us a bit more information as to what's going on.
On the client machine, please run a tail -n 500 /var/log/messages and post the output?
Can you please run the command over the CLI as requested above, and post the log for your /var/log/messages? We need more information to troubleshoot this.
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Tue Mar 01, 2016 7:41 pm
by rajasegar
dlukinski wrote:rajasegar wrote:Looks familiar.
You might be missing the glibc.x86 library.
Run the plugin manually from the server to see the error message.
Then ldd the plugin (if applicable) to see if there is any missing dependencies.
Could you help with listing what has to run (we have no access to these servers, everything done by the vendor)
- need some level of detail in the instruction to forward
Please
You need access to the server, or if you enabled debug the system log /var/log/messages.
Not sure exactly what plugin you used, so you need to look at nrpe.cfg and trigger the command manually in the server.
Assuming you are using check_load you can try this in the client server to see if there is any missing dependencies.
If there is any missing you can see it.
It could be a number of other things also.
Code: Select all
[nagios@nagiosprodxi1 libexec]$ ldd check_load
linux-vdso.so.1 => (0x00007fff3d3fc000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x00000034c1a00000)
libdl.so.2 => /lib64/libdl.so.2 (0x00000034c1200000)
libc.so.6 => /lib64/libc.so.6 (0x00000034c1600000)
/lib64/ld-linux-x86-64.so.2 (0x00000034c0e00000)
[nagios@nagiosprodxi1 libexec]$
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Wed Mar 02, 2016 10:53 am
by rkennedy
Thanks @rajasegar. @dlukinski - let us know when you get a chance to look at the previous responses.
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Thu Mar 03, 2016 1:37 pm
by dlukinski
rkennedy wrote:Thanks @rajasegar. @dlukinski - let us know when you get a chance to look at the previous responses.
Would it make sense to run ldd check_load if test versions of SuSe & CentOS (not Redhat unfortunately) work well if/when installed by us instead of trying to deal with servers managed by the vendor?
Example:
SuSe 11 SP3 or CentOS 6.x XI NRPE monitoring would work when I install and patch it.
When we follow same process (plus pre-reqs plus 2 x glibc updates on vendor-managed Linux, we get CPU, Memory and Open files checks in "unknown" state
-----------------------------------------------------
I'll provide RedHat results too in coming days
-----------------------------------------------------
Actually please close this thread, I am to open up new one
- latest results show same problems with RedHat ans SuSe in our case
Re: Many NRPE client checks do not work on REDHAT Linux
Posted: Thu Mar 03, 2016 3:34 pm
by rkennedy
Closing this thread out as requested, we'll continue in your new thread.