Ok, so we have ~100 AIX servers and ~300 Linux servers. Those numbers are goign to grow to ~200 and ~1000+. If I need to make a change to config, I don't want to have to update nrpe.cfg on all the servers and restart the service. That would just take forever and is just crazy to ask the admins to do that.
Am I able to use NRPE for the "standard" checks, like drives, mem, cpu, etc and then use NRDS to add anything extra? Meaning, I can use both on servers, right? What is best practice in this situation?
Thanks
NRDS and other checks
NRDS and other checks
2 of XI5.6.14 Prod/DR/DEV - Nagios LogServer 2 Nodes
See my projects on the Exchange at BanditBBS - Also check out my Nagios stuff on my personal page at Bandit's Home and at github
See my projects on the Exchange at BanditBBS - Also check out my Nagios stuff on my personal page at Bandit's Home and at github
Re: NRDS and other checks
I answered my own first question with an "of course" but I now have a second question.
I looked at the install script and see that it REQUIRES bash is installed. We do not have it installed on any of our AIX servers. Is there a reason you are requiring bash?
I looked at the install script and see that it REQUIRES bash is installed. We do not have it installed on any of our AIX servers. Is there a reason you are requiring bash?
2 of XI5.6.14 Prod/DR/DEV - Nagios LogServer 2 Nodes
See my projects on the Exchange at BanditBBS - Also check out my Nagios stuff on my personal page at Bandit's Home and at github
See my projects on the Exchange at BanditBBS - Also check out my Nagios stuff on my personal page at Bandit's Home and at github
Re: NRDS and other checks
You can use both agents just fine. Alternatively, many of our clients in dynamic environments push their nrpe configs with puppet/chef. So that is an option as well, though nrds offers much of the same functionality (though nagios specific).BanditBBS wrote: Am I able to use NRPE for the "standard" checks, like drives, mem, cpu, etc and then use NRDS to add anything extra? Meaning, I can use both on servers, right? What is best practice in this situation?
If you do end up using nrds extensively, it may make sense to migrate most of your checks to the nrds agent just for simplicity's sake as a single unified agent is usually easier to manage. Any of these scenarios are could be considered best practice depending on the environment.
Former Nagios employee
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
Re: NRDS and other checks
The send_nrdp portion of nrds uses a bash shell script (send_nrdp.sh) to send the checks back to nagios. Now there are a couple of other sending options, but understand that these are under-tested and may require some hacking depending on the environment. Your nrds config includes an additional setting for "send_nrdp", this is where the bash shell script is defined. You could try to change the sha-bang line to your shell, or prepend the the field with the full path to your shell, i.e.BanditBBS wrote: Is there a reason you are requiring bash?
Code: Select all
/bin/ksh /usr/local/nrdp/clients/send_nrdp.shThere was some work done a while ago to attempt to migrate the entire agent to perl, unfortunately, the person that was working on it is out of the office for the next few weeks. My apologies for the lack of a clearly defined solution, but maybe one of the above methods will work on your AIX boxes. By all means, if you hack out a solution, share it with us, but if you run into problems, remember that we are here to help.
Former Nagios employee
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.