Nagios Support Forum

I'm very newbie with Nagios and have spent the past couple weeks reviewing and learning about Nagios. We are deploying Nagios in AWS and would like to use Mod-Gearman for load-balancing. I reviewed the distributed monitoring options and it appears that Mod-Gearman will work with Core. I added the Nagios repo and installed gearman packages with yum on our instance at AWS. The gearmand package installs but always fails to start (the worker service installs and runs fine). Here are the steps and error received.

rpm -Uvh https://repo.nagios.com/nagios/7/nagios ... noarch.rpm
yum install gearmand gearmand-devel gearmand-server mod_gearman mod_gearman-debuginfo
Installed:
gearmand.x86_64 1:0.33-7 gearmand-devel.x86_64 1:0.33-7 gearmand-server.x86_64 1:0.33-7 mod_gearman.x86_64 0:3.0.7-1.el7
mod_gearman-debuginfo.x86_64 0:3.0.7-1.el7

Dependency Installed:
ncurses-compat-libs.x86_64 0:6.0-8.20170212.amzn2.1.3
boost-program-options.x86_64 0:1.53.0-27.amzn2.0.3

Has anyone had luck installing Mod-Gearman on Amazon Linux?

My second question is can we run a new Core version (we have 4.4.6 installed currently) with Mod-Gearman? I have read some old posts that Mod-Gearman won't work with Core after version 4.2.x but I am not finding a definitive answer as those posts are ~2 years old.

I appreciate any feedback and advice.

Todd

Just noticed I forgot to post the gearmand error...

gearmand: /usr/sbin/gearmand: symbol lookup error: /usr/sbin/gearmand: undefined symbol: _ZN5boost15program_options3argE

Apologies.

It isn't an answer to your question but I'd skip gearman until (or unless you) need it. I recently got rid of gearman on the last few hosts that had it configured. Same with the ramdisk.

Hi @toddf,

I think Gormank has a good point. Mod Gearman is great software but it also raises the overall complexity and time required to administer. I'd wait to integrate that until it's really needed.

Also, using passive services is another option here.

--Benjamin

I did finally get this working in AWS. Using docker containers to run both the gearman job servers and workers. It should allow us to scale up easily. We are still waiting on some firewall rules to test running checks on servers but having everything communicate was a good first step. I will post again on this topic once we get everything running.

I did finally get this working in AWS. Using docker containers to run both the gearman job servers and workers.

Nice!

Looking forward to your update.