Page 1 of 2

Putting mutiple hosts into downtime

Posted: Tue Feb 14, 2017 2:11 am
by Fred Kroeger
Recently upgraded to XI 5.4.2 and am unable to put multiple hosts into downtime from the Scheduled downtime page.
I can select multiple hosts OK. I select Trigger for Child Hosts and Select All services. After I apply it, only the first host shows on the "In Effect" table.
This was working OK on previous versions.

Regards... Fred

Re: Putting mutiple hosts into downtime

Posted: Tue Feb 14, 2017 11:06 am
by dwhitfield
I spoke to the devs about this. Nothing should have changed, and I'm not seeing this on my test box.

The devs are thinking Core may be the issue. From what version of XI did you upgrade? On what OS version are you running XI?

Re: Putting mutiple hosts into downtime

Posted: Tue Feb 14, 2017 5:33 pm
by Fred Kroeger
It's a very old VM that has been updated many times now so it's still running Postgres.
OS = CentOS 6.4 It was upgraded from XI 5.3.4 to 5.4.2
I've checked the test instance which is running 5.4.1 and it doesn't have this issue.
I've had a closer look this morning and created two hosts with a few services each and put both of the Hosts into downtime and selected All services.
The screenshot shows that 1 Host went into downtime with all it's services, but the second host only had the services go into downtime.
This is confirmed when I look at the Host Status page for this device- Host is not in Downtime but all it's services are.
downtime.PNG

Re: Putting mutiple hosts into downtime

Posted: Wed Feb 15, 2017 1:51 pm
by dwhitfield
I created bug report 11060 for this issue. Please refer to that bug # if you have anything else to add to the report. Did you have anything else to add at this point or should we lock it up?

Re: Putting mutiple hosts into downtime

Posted: Wed Feb 15, 2017 9:21 pm
by Fred Kroeger
Further testing today shows that the problem of the additional hosts not going in to downtime is only when you select Downtime for all Associated services.
If I select multiple hosts and don't select all the services, then the selected hosts go into downtime as expected.

Just went to tracker.nagios.com to update the bug report but I can't find bug report 11060 or any number that large?
Thanks ... Fred

Re: Putting mutiple hosts into downtime

Posted: Thu Feb 16, 2017 8:36 am
by dwhitfield
Tracker is deprecated. I am not 100% sure why. That's an internal bug #, but you can use it as a reference. However, in this case it should be fixed in 5.4.3.

In investigating the bug, the devs did find some other weirdness. Thank you for the additional info!

Re: Putting mutiple hosts into downtime

Posted: Mon Feb 20, 2017 8:42 pm
by Fred Kroeger
Some more weirdness to add to your Downtime investigations.
Recurring Downtime somehow got broken with the upgrade to XI 5.4.3.
It was working fine - Everyday Downtime starts at 18:45 and ends the next day at 07:15 - see the last 3 lines in the Notification list below.
Then another Downtime started at 2017-02-14 20:00:05. Instead of the Downtime ending at 07:15 the next day - it then started ending at 18:45. Then of course I started getting the notifications because the server is shutdown during the normal downtimes.
I checked the Recurring Downtime schedule and nothing had changed, it still had the original times. However I did notice that if I look at the comments for the Server itself, it shows the scheduled downtime start in my local time now (3 hrs behind the server time).
Regardless, the Recurring Downtime schedule shows that Downtime starts at 18:45 and ends at 07:15 the next day, but in fact the Downtime now Starts at 20:00 and ends at 18:45 the next day as per below.

I have deleted the Recurring Downtime schedule with all the scheduled downtimes and recreated the Recurring Downtime schedule. Looking at the comments, I can see that it is now scheduled and will check tomorrow to see if it is scheduling correctly

Code: Select all

Date / Time 		Host 	         Service 	Reason 	Escalated 	State 	Contact 	Dispatcher 			Information
2017-02-21 07:03:26	Server-X	-	Host Recovery	   	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.583ms, lost 0%
2017-02-20 19:14:55	Server-X	-	Host Problem	  	 No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-20 18:44:59	Server-X	-	Downtime Ended	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.104ms, lost 0%
2017-02-17 07:04:02	Server-X	-	Host Recovery	   	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.590ms, lost 0%
2017-02-16 20:00:05	Server-X	-	Downtime Started	No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-16 19:15:09	Server-X	-	Host Problem	   	No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-16 18:44:59	Server-X	-	Downtime Ended	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.095ms, lost 0%
2017-02-16 07:02:01	Server-X	-	Host Recovery	  	 No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.449ms, lost 0%
2017-02-15 20:00:06	Server-X	-	Downtime Started	No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-15 19:13:41	Server-X	-	Host Problem		No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-15 18:44:59	Server-X	-	Downtime Ended	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.144ms, lost 0%
2017-02-14 20:00:05	Server-X	-	Downtime Started	No	DOWN	Support	Custom: Email-host-event	CRITICAL - x.x.x.x: rta nan, lost 100%
2017-02-14 18:44:59	Server-X	-	Downtime Started	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.299ms, lost 0%
2017-02-14 07:14:59	Server-X	-	Downtime Ended	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.310ms, lost 0%
2017-02-13 18:44:59	Server-X	-	Downtime Started	No	UP	Support	Custom: Email-host-event	OK - x.x.x.x: rta 1.406ms, lost 0%

Re: Putting mutiple hosts into downtime

Posted: Tue Feb 21, 2017 10:16 am
by dwhitfield
Fred Kroeger wrote: Recurring Downtime somehow got broken with the upgrade to XI 5.4.3.
I assume you mean 5.4.2...or did someone give you a 5.4.3 pre-release? You are correct though, the devs did notice some other weirdness.

Please let us know what you find.

Re: Putting mutiple hosts into downtime

Posted: Tue Feb 21, 2017 5:56 pm
by Fred Kroeger
Yes you're correct of course it's 5.4.2 (upgraded from 5.3.3 )
On a random sample of 1 day..... Downtime started & ended at the correct time with the new schedule (created with exactly the same values as the old schedule).
Looks like something got screwed up during the upgrade?

Re: Putting mutiple hosts into downtime

Posted: Wed Feb 22, 2017 12:42 pm
by ssax
That is an internal bug report ID, it's not externally available, the developer has seen it and indicated there is an issue.

Please send the output of these commands as well:

Code: Select all

echo "select count(*) from nagios_scheduleddowntime where instance_id != '1';" | mysql -uroot -pnagiosxi nagios
echo "select count(*) from nagios_instances where instance_id != '1';" | mysql -uroot -pnagiosxi nagios