issues relating to nagios and lack of nrpe on nagios unit

[Jeff Hillman]

Stable charms
Bionic

When relating rsyslog-forwarder-ha to the nagios unit, an error appears:

---

2020-08-05 13:23:57 ERROR juju.worker.uniter.operation runhook.go:132 hook "config-changed" failed: exit status 1
2020-08-05 13:25:19 DEBUG config-changed Traceback (most recent call last):
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/config-changed", line 294, in <module>
2020-08-05 13:25:19 DEBUG config-changed hooks.execute(sys.argv)
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/charmhelpers/core/hookenv.py", line 909, in execute
2020-08-05 13:25:19 DEBUG config-changed self._hooks[hook_name]()
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/config-changed", line 287, in config_changed
2020-08-05 13:25:19 DEBUG config-changed update_nrpe_config()
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/config-changed", line 123, in update_nrpe_config
2020-08-05 13:25:19 DEBUG config-changed nrpe_compat.write()
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/charmhelpers/contrib/charmsupport/nrpe.py", line 301, in write
2020-08-05 13:25:19 DEBUG config-changed self.nagios_servicegroups)
2020-08-05 13:25:19 DEBUG config-changed File "/var/lib/juju/agents/unit-rsyslog-forwarder-ha-1/charm/hooks/charmhelpers/contrib/charmsupport/nrpe.py", line 195, in write
2020-08-05 13:25:19 DEBUG config-changed with open(nrpe_check_file, 'w') as nrpe_check_config:
2020-08-05 13:25:19 DEBUG config-changed IOError: [Errno 2] No such file or directory: '/etc/nagios/nrpe.d/check_rsyslog.cfg'

---

It appears the r-f-ha is expecting nrpe to be here. from the code:

---
@hooks.hook("nrpe-external-master-relation-changed")
@hooks.hook("local-monitors-relation-changed")
def update_nrpe_config():
    nrpe_compat = nrpe.NRPE()
    nrpe_compat.add_check(
        shortname="rsyslog",
        description="Check rsyslog is running",
        check_cmd="check_procs -c 1: -C rsyslogd"
    )
    nrpe_compat.write()

---

This works on all other units since we have nrpe-{host|container} there, but it is redundant and not sensible to have nrpe on a nagios unit.

running 'mkdir -p /etc/nagios/nrpe.d' on the nagios unit where r-f-ha is at resolves the issue.

[Adam Dyess (addyess)]

Yeah, seems like there's definitely an issue here.

[Drew Freiberger (afreiberger)]

I've seen this as well. This means that rsyslog service is not monitored on nagios units where nagios monitors itself (instead of using nrpe subordinate plugin).

This could be an issue for any subordinate charms that try to add_check. I'm going to subscribe the Nagios charm as I believe nagios should add this directory and include those checks in it's configuration somehow if it's set to monitor self.

[Eric Chen (eric-chen)]

This charm is no longer being actively maintained. Please consider using the new Canonical Observability Stack instead.
(https://charmhub.io/topics/canonical-observability-stack)
I will close this issue