StartAgents in agentd.conf weirdness

[Holgilein]

Hi all,

Today I stumbled into a strange effect, when I reduced StartAgents to "1"
and restarted zabbix_agentd on one of my virtual servers.

zabbix_server on my monitoring host started to throw these messages
afterwards:

032700:20060908:182514 Timeout while receiving data from [VS-250]
032700:20060908:182514 Getting value of [system.users.num] from host [VS-250] failed
032700:20060908:182514 The value is not stored in database.
...

When I restored the value back to "5", it works nicely again. Setting it to "3"
and zabbix_server starts throwing errors intermittendly (the lower the value,
the higher the error message density).

Furthermore, zabbix_server does flag the monitored server as "Not
Available" when StartAgents is set to "1", which lead me on a wrong trail
first, i.e. I started digging for a mis-configuration.

Any explanation would be very much appreciated.

Cheers.

[Alexei]

Number of agetns is basically number of simultaneous requests the agent may handle. If one agent's process is busy with something, it cannot process further requests.

[Holgilein]

Ahhh! That explains the strange behaviour, thanks!

You may probably want to add a comment to the config file, otherwise other
people may try to "optimise" the number of daemons running like I did...

PS: After another day of "horsing around" with Zabbix I am still very fond of it.

Goood stuff! (tm)