XI cannot send a notification email when service critical.

[dimsum]

Hi,

My XI worked after last week XI cannot send an email when host or service critical. Test send email is working. I have upgraded from 5.5 to 5.6 already but not help.

Any idea please advise.

Thanks.

[jacek]

Have exactly the same issue (but I was already on 5.6.0 when things stopped).
Stopped working for me somewhere between 24 and 25th.
We are using SMTP for sending mails and test mails are working without issues.
the eventman.log is completely empty.

I have an ongoing ticket about an other issue and I mailed in that ticket about this issue as well (#475721).

[tbs-infrastructure]

I am also facing the same issue. Despite NagiosXI showing that it has sent a notification, I got no emails, however, when testing the emails settings, I get the test email, so those settings are good. I am also on 5.6 and it was working fine until yesterday I had to restart the VM.

[jacek]

Looks like something might be outdated (like a certificate or so)?
I also rebooted the nagios server a few times, I expanded RAM and disk space.
One additional thing I noted was that the Performance grapher doesn't start after a reboot, need to do the start action - don't think it's related.

What do you see when you run this command and then force some immediate checks and custom notifications?
Nothing happens on my site (PROCESSED 0 EVENTS all the time)

Code: Select all

tail -f /usr/local/nagiosxi/var/eventman.log

[tbs-infrastructure]

Looks like something might be outdated (like a certificate or so)?
I also rebooted the nagios server a few times, I expanded RAM and disk space.
One additional thing I noted was that the Performance grapher doesn't start after a reboot, need to do the start action - don't think it's related.

What do you see when you run this command and then force some immediate checks and custom notifications?
Nothing happens on my site (PROCESSED 0 EVENTS all the time)

Code: Select all

tail -f /usr/local/nagiosxi/var/eventman.log

Same for me.

[lmiltchev]

Start a running tail on the event_handler.log:

Code: Select all

tail -f /usr/local/nagiosxi/var/event_handler.log

and watch the output for a while for errors. This may help identify the issue.

[tbs-infrastructure]

Here what I got:

tail -f /usr/local/nagiosxi/var/event_handler.log
(
[eventqueue_id] => 81
[event_time] => 2019-04-26 11:48:02
[event_source] => 2
[event_type] => 1
[event_meta] => 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
)
LOCKFILE '/usr/local/nagiosxi/var/event_handler.lock' STILL EXISTS - EXITING!
LOCKFILE '/usr/local/nagiosxi/var/event_handler.lock' STILL EXISTS - EXITING!
LOCKFILE '/usr/local/nagiosxi/var/event_handler.lock' STILL EXISTS - EXITING!
LOCKFILE '/usr/local/nagiosxi/var/event_handler.lock' STILL EXISTS - EXITING!

[tbs-infrastructure]

One more thing, this was caused by the update to 5.6, I created a snapshot of the VM before the update and reverting to the previous version everything starts working again.

[jomann]

Please delete the lock file:

Code: Select all

rm -f /usr/local/nagiosxi/var/event_handler.lock

This should fix the issue. The issue occurs when the event_handler cron does not properly stop. This could happen during restart. I've attached a new cron file that should fix the problem. It will also be fixed in XI 5.6.1.

To implement the fix, copy and overwrite the event_handler.php file in the /usr/local/nagiosxi/cron folder.

[tbs-infrastructure]

The issue seems to the tail -f /usr/local/nagiosxi/var/event_handler.lock file, by renaming that file the notifications started working again.