4.2.0 not reloading

[PitoneMaledetto]

Hi, as per Nagios Core Support Manager advise I am referencing this thread to address an issue that I am having on reloading the configuration on Nagios Core 4.2.0 (and previously on 4.1.1).
The reload does not work forcing me to restart nagios every time I make a change in configuration.
Please let me know if you need further details on top of those already stated on the forum thread linked above.
Thanks

[jfrickson]

After exchanging a few messages on the support thread, I'm switching here.

@PitoneMaledetto run the following commands as root:

rm /etc/systemd/system/multi-user.target.wants/nagios.service
rm /etc/systemd/system/nagios.service
systemctl daemon-reload

Then, edit /etc/init.d/nagios and starting at line 255, change this:

    reload|force-reload)
        if test "$checkconfig" = "true"; then
            printf "Running configuration check...\n"
            check_config
        fi

        if test ! -f $NagiosRunFile; then
            $0 start
        else
            pid_nagios
            if status_nagios > /dev/null; then
                printf "Reloading nagios configuration...\n"
                killproc_nagios HUP
                echo "done"
            else
                $0 stop
                $0 start
            fi
        fi
        ;;

to this:

    reload|force-reload)
        if test "$checkconfig" = "true"; then
            printf "Running configuration check...\n"
            check_config
        fi

        if test ! -f $NagiosRunFile; then
            printf "DEBUG - Running $0 start\n"
            $0 start
        else
            pid_nagios
            if status_nagios > /dev/null; then
                printf "DEBUG - killproc_nagios HUP\n"
                printf "Reloading nagios configuration...\n"
                killproc_nagios HUP
                echo "done"
            else
                printf "DEBUG - running $0 stop and $0 start\n"
                $0 stop
                $0 start
            fi
        fi
        ;;

Then do a tail -f on the nagios.log and run /etc/init.d/nagios reload and tell me which DEBUG line gets printed, and if nagios.log shows a reload happening.

[PitoneMaledetto]

Hi John,
I have performed all the changes as per instructions.
The reload still restart hence not reloading, I am not getting any DEBUG messages but I am getting this instead:

root@dev-nagios-01:/usr/local/nagios/libexec# /etc/init.d/nagios reload [....] Reloading nagios configuration (via systemctl): nagios.serviceWarning: Unit file of nagios.service changed on disk, 'systemctl daemon-reload' recommended. . ok

nagios.log
[1472721997] Caught SIGHUP, restarting... [1472721997] Event broker module 'NERD' deinitialized successfully. [1472721997] Nagios 4.2.0 starting... (PID=10034) [1472721997] Local time is Thu Sep 01 10:26:37 BST 2016 [1472721997] LOG VERSION: 2.0 [1472721997] qh: Socket '/usr/local/nagios/var/rw/nagios.qh' successfully initialized [1472721997] qh: core query handler registered [1472721997] nerd: Channel hostchecks registered successfully [1472721997] nerd: Channel servicechecks registered successfully [1472721997] nerd: Channel opathchecks registered successfully [1472721997] nerd: Fully initialized and ready to rock! [1472721997] wproc: Successfully registered manager as @wproc with query handler [1472721997] wproc: Registry request: name=Core Worker 11827;pid=11827 [1472721997] wproc: Registry request: name=Core Worker 11825;pid=11825 [1472721997] wproc: Registry request: name=Core Worker 11828;pid=11828 [1472721997] wproc: Registry request: name=Core Worker 11826;pid=11826

from the log I get a restart of the service as per [1472721997] Caught SIGHUP, restarting...

p.s. force-reload behaves the same.

[PitoneMaledetto]

Just a thought.
The status of systemctl is running (systemctl status) but I did not get anything back from the reload; I am not sure if the terminal should verbose a reload message on the terminal or not so I can't be 100% sure about the daemon reload.
Moreover systemctl stop nagios.service and systemctl start nagios.service do indeed work when executed.

[jfrickson]

It actually looks like it's working now. When the log says Caught SIGHUP, restarting..., it's not really restarting. It is doing a reload at that point.

I recommend you run the systemctl daemon-reload command, make a change in the nagios config, do a reload, and see if the change is reflected in the UI. I strongly suspect it will work.

[PitoneMaledetto]

Now I get:

root@dev-nagios-01:/usr/local/nagios/etc/objects/cy/servers# /etc/init.d/nagios reload [ ok ] Reloading nagios configuration (via systemctl): nagios.service.
So I assume it is normal that all the hosts get re-scanned too?

Because I remember that in previous versions of Nagios Core only the changes in configuration where PENDING.

[jfrickson]

In your nagios.cfg file, check the values of the retain_state_information, state_retention_file, retention_update_interval, use_retained_program_state, use_retained_scheduling_info, and all the retained_*_mask variables. I suspect one or more of them is set incorrectly.

[PitoneMaledetto]

Indeed it was:

Generic host definition

define host{
name generic-host ; The name of this host template
notifications_enabled 1 ; Host notifications are enabled
event_handler_enabled 1 ; Host event handler is enabled
flap_detection_enabled 1 ; Flap detection is enabled
process_perf_data 1 ; Process performance data
retain_status_information 0 ; Retain status information across program restarts
retain_nonstatus_information 0 ; Retain non-status information across program restarts
notification_period 24x7 ; Send host notifications at any time
check_command check-host-alive
register 0 ; DONT REGISTER THIS DEFINITION
}

In order to close this ticket would you say that removing:
rm /etc/systemd/system/multi-user.target.wants/nagios.service
rm /etc/systemd/system/nagios.service
and reload the daemon did the trick?

[jfrickson]

I would say yes. Most probably, those files were the cause.