Skip to main content

Hi,

I see

got no answer but it’s only 21 days old.

I initiate this thread because I’m also interested in any handy solution one could use to access to some aggregated statistics of this kind (ie: the whole total for the set of pollers/engines typically).

But also, and firstly, because, some weeks ago, on the 3 Centreon platforms I co-administer with my colleague, I noticed those charts were no more accessible, we had a message “there is no data” displayed briefly in every charts. Surprisingly, on the 3 platforms (which represents a total of 13 pollers, 10 if not counting the central servers), only one (external) poller still had been having the charts displayed.
I never took the time to troubleshoot the issue and hopefully, as you can see in the image below, upgrade to 23.10.115 fixed the issue. But only for 2 of the 3 platforms !

The 2 platforms which have been upgrade from 23.10.11 to 23.10.15 had the issue resolved.
The 1 platform which, for “administrative” reasons had been upgraded from 23.10.11 to 23.10.14, and then later  from 23.10.14 to 23.10.15, still has the issue (this is the platform which had the statistic working for a single poller, I can’t say if there is any kind of relation between those two facts, as for the slightly different upgrade paths. For the latter I’d hope it should be related, as doing the upgrade 11→14→15 versus 11→15 should be strictly equivalent, but I must recognize I doubt a bit. Let’s be more concise: I do not know what happen/happened, why it went “out of disorder” on two platforms but it hadn’t on the third one. I should watch closer (I suspect a file permission issue maybe ?) but currently I‘d appreciate any kind of hint or idea from any one.  This isn’t a critical issue though. I is working on the production platform and it is what matters the most.

NB: This is the second gap, on the right, that must be considered. The first one is caused by this platform to have been installed lately, and thus has no data before this date.

The three platform are now at version 23.10.16 but 15→16 didn’t fix the sh… thing!

Have a nice (and monitored!) day everybody.

RRD files for this charts seem to be there: /var/lib/centreon/nagios-perf

I checked the permissions and saw nothing wrong, they are updated.

I saw there is a tiny button to clear the data. I did it on our devel platform and charts have been recreated, tested also on the platform which has the problem but it changed nothing..

I suppose one can get a statistic like “how many services have been checked during the last hour (depends on the configuration of the RRD I guess), but it looks not easy to do so (at least to me).

Maybe there are some info at the BDD level which could be exploited for this?

Reply


Terms & ConditionsCookie settings