Nagios Support Forum

We're getting the same Stage 1 Report every minute.

I have already done these:

echo "truncate table xi_events; truncate table xi_meta; truncate table xi_eventqueue;" | mysql -h 127.0.0.1 -uroot -pYOURPASSWORD nagiosxi

/usr/bin/php /usr/local/nagiosxi/cron/dbmaint.php

How can I stop the deadpool emails?

Thanks.

Hi Greg,

I agree the steps you're taking should have cleared the cron queue so it's likely there is some bug with the way the deadpool script works that is failing to recognize that the object has been moved and is continuing to trigger the operation and corresponding notification. I'm curious what you see in your dbmaint.log file for this, as it may help us diagnose the issue.

Unfortunately my best suggestion right now for disabling the email is to turn off the deadpool processor, or remove the email address from your Deadpool Settings page. I have logged an issue for this component to address the issues with it, as I was struggling to get it to work in my own attempts to replicate the issue in the short term.

Hi Greg,

Your post has sparked an internal conversation regarding the deadpool. It seems that it works well enough for simple configurations, but with more complex configurations where a host or service is part of many different Nagios objects, it doesn't. Likely your configuration is complex enough to have crossed that threshold. We are considering removing the feature in upcoming versions.

Thanks for the post,

Aaron

Logs as requested, attached. This is with no email address in deadpool.

I have just re-added my email to deadpool, will wait for emails to start up again, send dbmaint.log for that as well.

I note this line in dbmaint.log: <p><pre>SQL Error [ndoutils] : Table './nagios/nagios_logentries' is marked as crashed and last (automatic?) repair failed</pre></p>

Investigating. Prob run a manual /scripts/reapairdatabase (or whatever) manually.

Since re-adding email, no events have come from deadpool. So something happend to clear the queue. Also nothing new in the dbmaint.log other that routine cleaning/repair complete every five minutes. Sure would like to get to a root cause. Event storms are disruptive.