NDO2DB service not starting after update to 5.7.5

[wbaars]

After updating from NagiosXI 5.6.14 to 5.7.5, NDO2DB cannot start:

Code: Select all

[root@redacted ndo]# systemctl status ndo2db.service
● ndo2db.service - LSB: Nagios NDO2DB Initscript
   Loaded: loaded (/etc/rc.d/init.d/ndo2db; bad; vendor preset: disabled)
   Active: failed (Result: exit-code) since Fri 2020-11-13 13:11:47 CET; 11min ago
     Docs: man:systemd-sysv-generator(8)
  Process: 13560 ExecStart=/etc/rc.d/init.d/ndo2db start (code=exited, status=1/FAILURE)

Nov 13 13:11:47 redacted.domain.ext systemd[1]: Starting LSB: Nagios NDO2DB Initscript...
Nov 13 13:11:47 redacted.domain.ext ndo2db[13560]: Executable file /usr/local/nagios/bin/ndo2db not found.  Exiting.
Nov 13 13:11:47 redacted.domain.ext systemd[1]: ndo2db.service: control process exited, code=exited status=1
Nov 13 13:11:47 redacted.domain.ext systemd[1]: Failed to start LSB: Nagios NDO2DB Initscript.
Nov 13 13:11:47 redacted.domain.ext systemd[1]: Unit ndo2db.service entered failed state.
Nov 13 13:11:47 redacted.domain.ext systemd[1]: ndo2db.service failed.
[root@redacted ndo]# /usr/local/nagios/bin/ndo2db --version
-bash: /usr/local/nagios/bin/ndo2db: No such file or directory

Seems like ndo2db is missing or not correctly installed. Any ideas?
The nagios.cfg seems to be updated correctly, with the new broker line

Also, Nagvis is not working anymore after the update.

Code: Select all

The backend with the ID "live_1" is not defined.

Changing the backend to "nagiosxi" does give the same error.

Code: Select all

The backend with the ID "nagiosxi" is not defined.

Any ideas?

[wbaars]

I've managed to fix the Nagvis issue:

Code: Select all

Found some info the CHANGES.txt:
- Fixed issue with NagVis on PHP 7.4 [TPS#15228] -JO
  (run "mv /usr/local/nagvis /usr/local/nagvis-back" and then run "/usr/local/nagiosxi/html/includes/components/nagvis/install.sh" to reinstall)

So, I executed the mentioned steps.

After I edited nagvis.ini.php to add some things and copied back our custom maps and iconsets to the /usr/local/nagvis directory.

Now Nagvis (1.9.8) is working as intended.

[tgriep]

XI version 5.7.x uses a new broker module that does not use the ndo2db daemon so when upgrading, it removed the ndo2db daemon as it is not needed.

[wbaars]

I see.
Why does the service ndo2db still exist then?
It also tries start automatically.
Is this something wrong with our upgrade?

[tgriep]

The upgrade does not remove the service check from the system so that is why it is still there.
It should of replaced it with a dummy check but you can delete it.

The upgrade should of removed the files that startup the ndo2db process but without any errors to look at, I don't know why.
As long as everything is working, you can delete the ndo2db service check in XI and disable it from starting.