Nagios Support Forum

Yesterday morning I migrated my Nagios installation from a VM on VMWare Workstation to a physical machine, and since doing so several of the remote servers I monitor have been intermittently giving the error "ERROR: Alarm signal (Nagios time-out)" when using check_snmp_win.pl to check the status of a Windows service. I am not sure if moving from VM to physical is the cause of the issue, but it seems a coincidence so thought it worth mentioning. Nagios is running on Centos using the same hardware as previously (albeit a different hard drive.)

If I run the plugin directly from the command line, I have no issue and it produces a result fairly quickly (3 or 4 seconds.) This is what I am running:

./check_snmp_win.pl -H <<IP Address>> -n "IIS Admin Service" -C public -r -t 60

I added the timeout as when Googling the error it seems this can help. I also set the Nagios service timeout to longer to see if this helped.

The server is using the same NIC and connection as previously, so I am not sure if it is due to moving from virtual to physical, or something else. I updated my check_snmp_win.pl script as well.

One other thing I have tried is to use the --v2c flag to use SNMP v2. The odd thing is that sometimes this gets a correct result quicker (when using command line), and sometimes it fails. This is the only time I can get "ERROR: Alarm signal (Nagios time-out)" to show up from command line. However, if I remove the flag, it will always work.

Here is an output of commands run one after the other:
So SNMP v1 always works on command line (when I have tested), v2 either works quicker or gives the error, and both can give the error through Nagios.

I am running out of ideas of what could be the cause, and what the solution may be. Any help would be much appreciated, I have only been using Nagios for a week or so, so I am pretty new to it.

Did the IP address or FQDN change when you migrated?

If you go to http://YOURSERVER/nagiosxi/admin/globalconfig.php are the Program and External URLs still correct? The External can be blank, as long as XI does not need to be accessed from outside your network.

dwhitfield wrote:Did the IP address or FQDN change when you migrated?

If you go to http://YOURSERVER/nagiosxi/admin/globalconfig.php are the Program and External URLs still correct? The External can be blank, as long as XI does not need to be accessed from outside your network.

Hi, thanks for responding, no the IP address and FQDN both stayed the same when I migrated. It is odd because the checks can work fine for a while then sporadically give the error with no pattern.

I don't see anything at that link, but notice is contains nagiosxi and I am running core.

I have actually started using centreon_plugins.pl this afternoon, with the "--plugin=os::windows::snmp::plugin --mode=service" switch, and that looks like it does the same thing but is more reliable at the moment, so I may well just move over to that (from check_snmp_win.pl).

If that works fine I will come back and mark this as resolved.

Chapzele wrote: I don't see anything at that link, but notice is contains nagiosxi and I am running core.

Ok, then the place to look other than the places you mentioned would be the httpd.conf or whatever it is on the distro you are running...but I really doubt that's the issue.

Let us know how the new plugin goes! There's no real way for you to mark it as resolved other than editing the title in the first post, which you are welcome to do, or you can just leave a note telling us it's resolved.