Centreon: Centreon 18.10 Ack/downtime/External commands fails

[lencell]

---

BUG REPORT INFORMATION

Centreon Web version: 18.10.1

Centreon Engine version: 18.10.0

Centreon Broker version: 18.10.1

OS: CentOS Linux release 7.6.1810 (Core)

Additional environment details (AWS, VirtualBox, physical, etc.): Hyper-V Virtual Machine, installed from the ISO centreon-18.10-5.el7.x86_64 available on https://download.centreon.com

Steps to reproduce the issue:

1. Select a Host or a Service in the Monitoring menu
2. Select acknowledge, set downtime or any of the other choices
3. Validate the box

Describe the results you received:
Nothing happens

Describe the results you expected:
Service or Host, Ackowledged/downtimed/any of the other choice applied

Additional information you think important (e.g. issue happens only occasionally):
Happens 100% of the time, never worked since the install (yesterday the 20th of December)
I had this issue occasionaly on pollers of previous versions so I checked that the file /var/lib/centreon-engine/rw/centengine.cmd exists, is a valid FIFO and belong to centreon-engine, /usr/lib64/centreon-engine/externalcmd.so is also properly configured in the engine configuration panel. Everything seems ok
Lastly the call from the web page seems properly logged in httpd access log:

[21/Dec/2018:10:28:55 +0100] "GET /centreon/include/monitoring/external_cmd/cmdPopup.php?cmd=72&comment=Acquitt%C3%A9%20par%20someuser&sticky=true&persistent=true&notify=false&ackhostservice=true&force_check=true&author=someuser&select[centreon-central]=1&select[somemachine.tld]=1 HTTP/1.1

I found nothing more related to this in the various log files (/var/log/{centreon,centreon-broker,centreon-engine}/*.log)

I can provide more information if requested

[StefThomas]

I can’t reproduce… Could it be a permission issue? Did you try with an account which has administrator privileges ?

If it’s not already done you should activate the debug in your poller’s engine configuration, you’ll get more info than in the usual log.

[sc979]

Hi lencell,
1- Can you tell us which language you use ?
2- Can you send us the external command sent :
tailf /var/log/centreon-engine/centengine.log | grep DOWNTIME
when you add a downtime.
3- And can you send us the php7 log in :
/var/opt/rh/rh-php71/log/php-fpm/centreon-error.log

Thanks in advance.
Best Regards,
sc979

[lencell]

Hi, thanks to both of you for your answers.

I tried with the admin account that is created with the initial setup (although I renamed it so it's not "admin")

I tried with the following languages:

• fr_FR.UTF-8
• en_US.UTF-8
• en_US

cat /var/log/centreon-engine/centengine.log | grep -i DOWNTIME
Gave no output, (I tried to set a downtime and run it after)

the php7 log is only 16 lines long, which contains the 3 following errors:

[21-Dec-2018 10:31:58 Europe/Paris] PHP Notice:  Undefined index: centreon in /usr/share/centreon/www/include/monitoring/external_cmd/popup/popup.php on line 49
[21-Dec-2018 10:31:58 Europe/Paris] PHP Notice:  Undefined index: HTTP_ACCEPT_LANGUAGE in /usr/share/centreon/www/class/centreonLang.class.php on line 135
[21-Dec-2018 11:44:26 Europe/Paris] PHP Notice:  Undefined offset: 1 in /usr/share/centreon/www/index.php on line 112

As for the debug mode for the poller, I tried it right after I opened this issue, by checking log everything with the highly detailed option but it got me nothing interesting

Best Regards,

[lpinsivy]

First edit /etc/sysconfig/centcore and change "error" to "debug"
Second go to "Administration > Parameters > Centcore" menu and enable debug
And Restart centcore process:

systemctl restart centcore

Then can you check in /var/log/centreon/centcore.log if you can see your downtime?

[lencell]

Restarting centcore did the trick, I think the problem is that it was not started in the first place and it's the only thing I didn't checked.

So no bug, just a problem between the chair and the keyboard, although if I can suggest a change for your next release of CES: enabling by default the services cbd, centengine and centcore, as the first two were not started either (after a reboot or otherwise) and caused me a bit of trouble too.
You probably have the commands already although I will still put them here for reference:

# Enable autostart
systemctl enable cbd
systemctl enable centcore
systemctl enable centengine
# Start the services
systemctl start cbd
systemctl start centcore
systemctl start centengine

Thanks a lot to everyone for your support.