Nagios Plugin for Cacti (NPC) 2.0.4

[sterpstra]

Yes, I'm. After making the changes you posted and the PDO change it's all working now.

Still no joy on this end. I made the changes you suggested but on top (no pun intended) of having a CPU spike when accessing NPC, I still only get the dashboard partially populated... VERY strange....

-st

[mati]

I'm using the svn version and I only made those 2 changes. I have config_output_options=2 in ndomod.cfg and retain_state_information=1 in nagios.cfg other than that everything else is set to default and the recomended settings in the NPC quickstart guide.

[sterpstra]

Hi Divigater

Any update on this issue? I still can't get past the front page when I make any changes to the various config files.

-st

[sterpstra]

I finally found the source of the problem..... The Cacti config.php file is calling require('/etc/cacti/debian.php'); which had all of the database variables required by NPC with the exception of hostname and port, which where blanked out. After I changed those, NPC came up fine.

However, now I'm back to the original problem I had when I first was able to force a connection. I get the Dashboard page and the current status of Nagios is shown. However, no hostgroups or servicegroups are shown and when you go to those tabs (Hosts and Services), there is nothing there. Just doesn't make sense that it can pull some information from that db but not other info. Everything is back to the original code, so that shouldn't be the issue....

Ideas anyone?

[sterpstra]

And here's my new /etc/cacti/debian.php code:

$database_username='cacti';
$database_password='password';
$basepath=' ';
$database_default='cacti';
$database_hostname='localhost';
$database_port='3306';
$dbtype='mysql';

[xefil]

Does it work on Nagios 2.9?

[lewise.liu]

hi guys
Recently I have been busy in deploying the cacti+nagios monitor system.But at npc 2.0 model,I meet a odd error:
[1215233124] ndomod: Still unable to connect to data sink. 2297621 items lost, 5000 queued items to flush.(in nagios.log).
Cacti can works well ,and nagios can works normally also.But the NPC web is blank and has none data on the hosts and services.
At the same time ,I find the table Nagios_hosts and Nagios_services have rows. which is so odd.
The upper error happen after I deployed the ndomon and ndo2db to nagios.
My deploying version be as below:
Cacti 0.8.7b
Cacti Plugin Architecture v2.0
NPC 2.0
ndoutils-1.4b7
php 5.2
mysql 5.0
nagios-3.0
Perhaps somethings is wrong with the ndoutils.But I have reinstalled and deployed the ndoutils for 3 times. The problem exists also.
The problem has suffer me for one weeks.
Any suggest and help is appreciated.

[cixelsyd]

Hello all.

I am running nagios 3.0.1 with cacti and NPC. I get these NPC errors inside my httpd logs constantly:

[Wed Jul 16 09:07:15 2008] [error] [client ::1] PHP Warning: Variable passed to each() is not an array or object in /var/www/html/lib/functions.php on line 147
[Wed Jul 16 09:07:16 2008] [error] [client ::1] PHP Warning: reset() [<a href='function.reset'>function.reset</a>]: Passed variable is not an array or object in /var/www/html/lib/functions.php on line 146
[Wed Jul 16 09:07:16 2008] [error] [client ::1] PHP Warning: Variable passed to each() is not an array or object in /var/www/html/lib/functions.php on line 147
[Wed Jul 16 09:07:16 2008] [error] [client ::1] PHP Warning: reset() [<a href='function.reset'>function.reset</a>]: Passed variable is not an array or object in /var/www/html/lib/functions.php on line 146


It seems like most things work; however, inside NPC when I click services or hosts I get a blank screen ... almost like the query is timing out before it can return values.

Any ideas?
Cixel

[nduda78]

Any reason why this plugin breaks the refresh of everything else in cacti (normal graph pages, weathermap...etc)

The source code says this on any page now:

meta http-equiv=refresh content='2592000'

Disabling NPC fixes everything.

[petero]

Hi
I cant seem to get NPC tab get to show up, i can't no errors.
And there is no "realm" stuff in setup.php (with new version its 9000+x)
so there is no conflict with realm thing.
Any one got it to show up ?

TW:
Full url work:) like /cacti/pluginc/npc/npc.php

Peter

[divagater]

nduda78, you are right that the huge meta refresh is making it into other pages. To me that looks like a plugin architecture issue. Each plugin developer can set the refresh for their plugin with the following code in setup.php:

Code: Select all

/**
 * Set the page refresh very high
 */
function npc_top_graph_refresh() {
    return('2592000');
}

Because NPC is a client side application it does not need to ever refresh so I set the value very high. What is interesting is that when you navigate away from NPC to another tab that value seems to be carried over. The quick fix if you want refresh working is to comment out that method or set the value low. But you will not like it when NPC refreshes and all your tabs close.

I will see if I can find out why the value does not reset when navigating away from npc.

Cheers,
Billy

[TheWitness]

I suggest you create your own top_graph_header.php, instead call it top_npc_header.php and remove the refresh bit.

TheWitness

[daldworth]

I Keep getting these messages:

Code: Select all

[Sun Aug 10 22:17:23 2008] [error] [client 63.211.......] PHP Fatal error:  Uncaught exception 'Doctrine_Connection_Exception' with message 'Couldn't locate driver named mysql' in /usr/local/share/cacti/plugins/npc/lib/Doctrine/Connection.php:394\nStack trace:\n#0 ...

Yet I know for certain mysql is loaded into php. Cacti is using it at least. And:

Code: Select all

# php -m
[PHP Modules]
date
json
libxml
mysql
pcre
PDO
Reflection
session
SimpleXML
snmp
sockets
SPL
standard
xml

Otherwise all else is good. NDO2DB data is loading from a remote central nagios server. NCSA is populating correctly. The NPC installl went all as planned except for this...

I'm on a FreeBSD 6.2 machine if that makes a diff.

Thanks in advance for any help.

[daldworth]

Figured it out. Needed pdo_mysql port installed... doh!

[sfxdana]

Excelent Plugin!! Keep up the good work!!!

Just started using Nagios for a few weeks now to monitor our customers, not using the NPC plugin jet. But will be in another few weeks or so.

Just one question/request is it possible to sort the service groups and host groups on there critical/warnings etc.

Just like the Olympic Medal country list. Sort on Criticals first then sort on warnings then ok messages. That way only the problems are displayed (top ten bassicly) and you don't have to go through all the pages to find the criticals/warnings etc.

Just have to look at the status page and know where the problems are. And act on them accordingly. Refresh time could also be set to less then 60 seconds so you see problems quickly.

Sfxdana