Nagios Support Forum

Gearman Packages currently installed on Main Server (gearman worker not running however):
gearmand-server-0.33-2.x86_64
mod_gearman2-2.1.1-1.el6.x86_64
gearmand-devel-0.33-2.x86_64
gearmand-0.33-2.x86_64


Workers all have
gearmand-devel-0.33-2.x86_64
gearmand-0.33-2.x86_64
libgearman-1.1.8-2.el6.x86_64
mod_gearman-1.4_nagios4-1.el6.x86_64
mod_gearman2-2.1.1-1.el6.x86_64


Processes are running:
/usr/bin/mod_gearman2_worker -d --config=/etc/mod_gearman2/worker.conf --pidfile=/var/mod_gearman2/mod_gearman_worker.pid






These values were changed to the below values:
max_execution_time = 120
max_input_time = 120
memory_limit = 512M

Did you add this to the /etc/php.ini file? If not add it.

Also, change the memory limit to the following
memory_limit = 1024M

Save the file and restart the following Did you do the changes in this link?
https://support.nagios.com/kb/article/n ... anner.html

Is the Apply Configuration working for you now?

Yes. max_input_vars=5000 was added.

I have also just changed memory limit to 1024M

I also made the changes to reflect:
for i in {1..40} I had 1..30 before

Is the Apply Configuration working for you now?
If not, can you login to the XI server as root and run this command, Then login to the XI GUI and Apply the configuration and post the output of the tail command.

On my DR box (manual DR)

I added 6 hosts. and I am seeing the checks going back and forth between being ok and Down with the status info of (host check orphaned, is the mod-gearman worker on queue 'host' running?)

Logs show alot of these:

[1505244751] Warning: The check of service 'Total Processes' on host 'server' looks like it was orphaned (results never came back; last_check=1505243890; next_check=1505244070). I'm scheduling an immediate check of the service...


/var/log/mod_gearman2/mod_gearman_neb.log shows:

[2017-09-12 05:00:03][46277][ERROR] sending job to gearmand failed: connect_poll(Connection refused) getsockopt() failed -> libgearman/connection.cc:104
[2017-09-12 12:09:02][29186][ERROR] sending job to gearmand failed: connect_poll(Connection refused) getsockopt() failed -> libgearman/connection.cc:104

BTW the prod server seems to be stable for now.
This DR box however doesn't seem to be.

I did some searching on that error and from what I found, it it caused by the gearmand server not listening for any new connections and that it is a possible bug in the software.
You can try and upgrade the gearman software or enable debugging in the gearmand module.conf file and see if there is more information in the log file on why it is failing.

I tried upgrading. Following https://assets.nagios.com/downloads/nag ... ios_XI.pdf

I noticed that when I did the server upgrade first and then the worker upgrade that the worker upgrade removed components of the server. So I had to do a straight install and eventually all of the packages were available.

gearmand-0.33-2.x86_64
gearmand-devel-0.33-2.x86_64
mod_gearman2-2.1.1-1.el6.x86_64
gearmand-server-0.33-2.x86_64

I seem to be at the latest version. Is this not the case?

After restarting gearman with debug mode on. It seems to have cleared everything up. Not sure how long this will last however. I will keep you posted.

Those versions are the last known versions that were tested on a Nagios XI server.
If you go to the Gearman web site, there are newer versions available but we do not have any instructions for installing the newer versions.