Incoherent information between zabbix_agentd AND zabbix_get

[nerique]

Hi all,

and before evrything, thanks for your help.

Here's my issue.

My Zabbix server is now configured, and it's monitoring one distant host.

Everything is running fine except i got a Apache is not running trigger raised, and a graph showing 0 processes Apache on my host.

When executing a ps aux | grep apache2 on my host, it returns 12 running processes.

When executing on the host zabbix_agentd -t proc.num[apache2], it returns 12
When executing on the server zabbix_get -sMYHOST -p10050 -k"proc.num[apache2]", it returns 0

Whant can the problem be? There is no error at all in both server and host logs.

I've gone through all the internet and all the documentation, and I can't find any solution.

Can anyone help please?

Thanks!

[nerique]

I have tested 1.8.3, 1.8.4rc2 and 1.9 versions of Zabbix and always the same thing.

Apache, Mysqld, etc...

it still returns 0 from the zabbix_get.

Anyone can help?

I can't find any solution

[nerique]

I think I'm able to figure out what's wrong, but i'm unable to solve it.
If i Run from server the command
zabbix_get -smboost-dev.m-boost.net -k"proc.num[]"
It returns 6, which is the number of actual running processes on host for user zabbix.

Can it be a unix user relatedf issue? User zabbix can't see others user processes.

How can I fix it?

Thanks!