Migrating Nagios XI host to CentOS 7

[rferebee]

Good morning,

I will be migrating my XI host to CentOS 7 this morning, but I'm running into a problem whenever I light up the new server.

Despite having notifications disabled in both the config file /usr/local/nagios/etc/nagios.cfg and in the GUI the new server is sending out dozens of "false positive" notifications. I assume they are cached somewhere?

My question, is there a way to clear out the notification cache on my new server so it won't send out anything during the transition period? I only want notifications to resume once the environment has stabilized and all of the hosts/services have checked-in.

Thank you.

[benjaminsmith]

Hello @rferebee,

Make sure you have the notification turned off.

disable_notifications.png

You should be able to view the mail queue with the following command, and can clear out the /var/spool/mqueue/ directory to remove message from this queue.

Code: Select all

mailq

[rferebee]

So, I was reading another thread and it mentioned the xi_events table.

Do the notifications have anything to do with that table?

[scottwilkerson]

Yes, if you had queued events they would still send

You can run the following to clear them all

Code: Select all

echo "truncate table xi_events; truncate table xi_meta; truncate table xi_eventqueue;" | mysql -u root -pnagiosxi

[rferebee]

I don't see /var/spool/mqueue is that bad?

[rferebee]

Yes, if you had queued events they would still send

You can run the following to clear them all

Code: Select all

echo "truncate table xi_events; truncate table xi_meta; truncate table xi_eventqueue;" | mysql -u root -pnagiosxi

I tried running this command, but it's giving me the following error:

Code: Select all

ERROR 1046 (3D000) at line 1: No database selected

[benjaminsmith]

Hello @rferebee,

Looks like the db got cut off, run this:

Code: Select all

echo "truncate table xi_events; truncate table xi_meta; truncate table xi_eventqueue;" | mysql -uroot -pnagiosxi nagiosxi

[rferebee]

Ok, that worked, but the host is still showing 9770 requests when I run the: mailq command

[benjaminsmith]

Hi @rferebee,

I believe you are using Cent 7? If so, likely using the postfix MTA. Run the following and let me know if that clears the queue.

Code: Select all

postfix flush

[rferebee]

I don't think sendmail is actually installed on this new CentOS 7 host. Which is probably why it's not behaving like it typically would.

The location where the sendmail configuration files should be /etc/mail doesn't exist on the new host. What's more confusing for me though is trying to figure out how the host is sending out notifications if sendmail isn't installed.

The postfix flush command didn't do anything. The queue just keeps growing. It's up to over 11k now.