Problem Backend:nagiosxi NDO claims Nagios did not update

[gaguilard]

Hi:
I am looking for a solution for blue icons in NagVis.
I has read a lot and is not clear for me the fault which makes Nagvis displays blue icons.

It situation ocurrs twice a day, but last days in the morning, during our monitoring meeting. This is a new behaviour because it never happened before, since two weeks.

The complete error is "Problem (Backend:nagiosxi) NDO claims Nagios did no status update from more than 60 secs. make sure that Nagios and NDO daemons are running. ( Bakend.ID:nagiosxi)

We are using the VM provided and updated by GUI update. Current NagiosXI version 5.8.5.

I understand that this error occurs when Nagvis is trying to read the database and Nagios do not updated it. But why Nagvis can't test if the db is updated, and if not, continue showing the last valid status, and showing a message that the data must not updated in the last try due data can´t be refreshed.

Please, can you help me with it matter?.

[ssax]

I can see a long apply configuration exceeding the 60 seconds default timeout and causing that.

Edit this file:

Code: Select all

/usr/local/nagvis/etc/nagvis.ini.php

Under this section:

Code: Select all

[backend_nagiosxi]

Change this:

Code: Select all

;maxtimewithoutupdate=180

To this:

Code: Select all

maxtimewithoutupdate=180

See if that helps alleviate it.

[gaguilard]

Hi ssax:
Thanks for your reply.
I did it few days ago and I changed from 180 to 60 and the result is not too different.
Is there possibilty to set Nagvis to not update the screen when data is not available?. I understand that Nagvis detects that the database is being filled, but the people that looks the screens ( users) no.
Is there a possiblity to set the screen refreshment after the database has filled?.
At this time I had Nagvis with blues till for ten minutes each time, but not the rest of times.
Then, if there is some process that reconstruct completely the data some times at day?.Is it possible to set this process to run it out of times that people make more use of this screens?.

Thank you.

[ssax]

Try changing it to 300 and see if that helps. That is the only option I see in Nagvis to adjust this.

In the Nagvis menu, go to the Options > General Configuration menu and increase the refresh time to a larger value as well.

[gaguilard]

Hi ssax:

Thanks again for your reply.

I changed it but is the same, at any time you will have a db update that avoid Nagvis has the data to present.

Another think to test?.

Thank you.

[ssax]

The only way I could get maxtimewithoutupdate setting to work for the backend was to edit this file:

Code: Select all

/usr/local/nagvis/share/server/core/classes/GlobalBackendPDO.php

Comment out this section (around line 114):

Code: Select all

            // Check that Nagios reports itself as running
            if ($nagiosstate['is_currently_running'] != 1) {
                throw new BackendConnectionProblem(l('nagiosNotRunning', Array('BACKENDID' =>$this->backendId)));
            }

So it looks like this:

Code: Select all

            // Check that Nagios reports itself as running
            //if ($nagiosstate['is_currently_running'] != 1) {
             //   throw new BackendConnectionProblem(l('nagiosNotRunning', Array('BACKENDID' =>$this->backendId)));
            //}

Then try again.

EDIT: I have submitted a bug report for this per development's request:

Code: Select all

NEW TASK ID 15606 created - Nagios XI Bug Report: XI - Nagvis - Nagvis not honoring maxtimewithoutupdate on apply configuration

[gaguilard]

Hi ssax:

I taked some days of vacation and here I am again. Please, my appologies for not post the reply before.

I made the changes sugested and we will have the Nagvis dasboards under observation to know the result.

I will let you know about results results in a few more days.

Thank you for your support

Kind Regards

Gastón

[ssax]

Sounds good, we'll keep an eye out for your update.

[gaguilard]

Hi ssax:

We tested changing the NagVis graph udate and the NDO update time, but without success.

We are thinking to change the backend but in my thougts it nothing change, because it is a matter of time that Nagvis find the backend making the fillup of the database, but, we will test this too.

We are investigating the way to do that.

Thank you.

Kind regards.

Gastón

[ssax]

I labbed it up before I sent it, trying changing the nagvis.ini.php backed for nagiosxi setting to this:

Code: Select all

maxtimewithoutupdate=300

Then that change should work if you've commented it out and will use the cached data up to 300 seconds (or whatever you set).