Error: no output from backend!
Posted: Thu Jul 22, 2021 4:18 pm
Hello,
I have a Nagios XI instance that's run into some difficulty. It's been under heavy load for a while. The last time I got assistance with it, it got its databases all converted to InnoDB. This helped immensely to stop the table crashes, but something else has happened. Now symptoms include:
Monitoring Engine Event Queue dashlet only says "Error: no output from backend!"
Hosts are displayed in the web UI, but no services are. This extends to the API as well:
Configuration changes still appear to be working. There was an old install of Check MK, which still shows the services. I was able to deactivate a couple checks, apply the updated configuration, and see the checks disappear from the Check MK page.
To further complicate things, there was a failed attempt to upgrade from 5.8.4 to 5.8.5. The last few lines output were:
The ^[[D were from me hitting keys to see if it was still alive. I had let the upgrade script run overnight but it never finished.
To even further complicate things, I moved the databases onto an external server to reduce the CPU load on the Nagios XI box. The doc warns about there being any errors, but the databases actually migrated just fine. The load on the Nagios XI box is much lower and putting the database on another VM seems to have allowed it to keep up with the check processing load. It's just that the service check results aren't showing up in the XI web UI or API (but like I mentioned, they're up-to-date as shown in the old Check MK web UI).
Thanks!
I have a Nagios XI instance that's run into some difficulty. It's been under heavy load for a while. The last time I got assistance with it, it got its databases all converted to InnoDB. This helped immensely to stop the table crashes, but something else has happened. Now symptoms include:
Monitoring Engine Event Queue dashlet only says "Error: no output from backend!"
Hosts are displayed in the web UI, but no services are. This extends to the API as well:
Code: Select all
[root@*** scripts]# curl --insecure -XGET "https://***/nagiosxi/api/v1/objects/servicestatus?apikey=****&pretty=1"
{
"recordcount": 0,
"servicestatus": [
]
}
To further complicate things, there was a failed attempt to upgrade from 5.8.4 to 5.8.5. The last few lines output were:
Code: Select all
UPGRADE: POST-UPGRADE: NDO post upgrade started...
Performing install...
> Enter MySQL root password: > Creating database with user
^[[D^[[D^[[C^[[D^[[D
> Press <ENTER> to continue... > Using command line supplied credentials for account/db creation...
> Username: ndoutils
> Password: *****
> Database: nagios
ERROR 1045 (28000): Access denied for user 'root'@'localhost' (using password: YES)
Something went wrong creating database 'nagios'To even further complicate things, I moved the databases onto an external server to reduce the CPU load on the Nagios XI box. The doc warns about there being any errors, but the databases actually migrated just fine. The load on the Nagios XI box is much lower and putting the database on another VM seems to have allowed it to keep up with the check processing load. It's just that the service check results aren't showing up in the XI web UI or API (but like I mentioned, they're up-to-date as shown in the old Check MK web UI).
Thanks!