Nagios Support Forum

Hello,

I have run into a recurring problem with Fusion. After X time (probably about a week), No one is able to login. The error is in the subject. I have found posts about using /usr/local/nagiosfusion/scripts/truncate_polled.php. This does fix the issue and accounts are able to login. The problem is that this keeps happening. Under system settings, i have changed log data retention to 1 day... does not help.

How can i stop this from happening?

Fusion version 4.1.8

Thanks

I wonder if the passwords are getting mangled in the database for some reason.

Is the server running out of space? Please also let us know if you see anything in your Apache or database log.

Hello,

Yes the / partition slowly fills up. I do not see any error in those logs since the lastime i rebooted the server. I could not find mysqld.log, I assume you meant mariadb.log?

You mentioned mangled passwords.. not sure if this is part of the overall problem, but the automatic passthrough from fusion to xi is not working either.

Thanks

Yes! Forgive me, I shouldn't have assumed you had a MySQL database.

If you see that the database is broken, you will probably want to run a repair. What do you see in your subsystem logs? You could tail them and then perform a test, or just send us all of the logs in that folder. They may tell us what's going on. FUSION SUBSYSTEMS

Can you also please check to make sure cron is running on that system?

Hello,
The / partition was full again this morning causing the issue again. I attempted to run a repair and teh following was the result.
Also: ps -ef | grep cron brings back entries.

I am working on my setup of a prod environment. I had this issue while testing and now that i am adding more to the environment it is crashing every couple of days.

What should be cleaning this up?

I downloaded fusion as a vm.

Looking at the subsys log directory the one that stands out is the dbmaint_subsys.log which had a different owner and a zero size file....

Okay, you will probably want to start up MySQL again and restart the cron service just to be safe. Can you please share ( PM ) your subsystem logs with us so we can try to determine what went wrong?

In addition to what @jbrunkow posted (please follow it), go to Admin > System Settings > Data & Polling tab and check the box next to "Only poll data for mapped users when a user is logged in" and see if that helps.

jbrunkow wrote:Okay, you will probably want to start up MySQL again and restart the cron service just to be safe.

Code: Select all

service mysqld start
Service crond restart

Can you please share ( PM ) your subsystem logs with us so we can try to determine what went wrong?

Code: Select all

/usr/local/nagiosfusion/var/log/*_subsys.log

Error starting mysqld

service mysqld start
Redirecting to /bin/systemctl start mysqld.service
Failed to start mysqld.service: Unit not found.

I will pm the logs shortly.

Thanks

ssax wrote:In addition to what @jbrunkow posted (please follow it), go to Admin > System Settings > Data & Polling tab and check the box next to "Only poll data for mapped users when a user is logged in" and see if that helps.

Thanks for the suggestion, I will try it in the morning.

ssax wrote:In addition to what @jbrunkow posted (please follow it), go to Admin > System Settings > Data & Polling tab and check the box next to "Only poll data for mapped users when a user is logged in" and see if that helps.

This morning, i enabled the only poll data for mapped users. ..

The "Log Server Alert History" dashlet stops updating with this setting on...

Had to turn it back off..