Zabbix agentd 1.1 won't start

[ReefShark]

I've been migrating from 1.1beta8 to 1.1 this week and on linux (SLES9), I've hit a snag; the freshly compiled agentd won't start.

# /etc/init.d/zabbix_agentd start
Starting zabbix_agentd startproc:  exit status of parent of /opt/zabbix/bin/zabbix_agentd: 1
                                                                     failed

I've turned logging on, and the logfile is created, but there are no entries. Compilation gave no problems at all; no errors, no warnings. The server is working fine on version 1.1, but so far, no luck on compiling the agents under linux. We have a few old RedHat 7 boxes, these give the same problem.

My configure settings:

./configure --prefix=/opt/zabbix \
--sysconfdir=/etc/zabbix \
--mandir=/usr/share/man \
--enable-agent

Other possible useable info:

Configuration:

  Detected OS:      linux-gnu
  Install path:     /opt/zabbix
  Compiler:         gcc
  Compiler flags:   -g -O2
  Linker flags:     -lresolv -lnsl 
  Compilation arch: linux

  Enable server:    no
  With database:    no

  Enable agent:     yes

My agentd config file:

Server=172.19.0.223
ListenPort=10050
StartAgents=5
DisableActive=1
DebugLevel=3
PidFile=/var/tmp/zabbix_agentd.pid
LogFile=/tmp/zabbix_agentd.log
Timeout=3
###############################################################################
# USER-DEFINED MONITORED PARAMETERS
###############################################################################

### System ###
UserParameter=system.cpucount,grep processor /proc/cpuinfo | wc -l

What am I missing here?

[fips]

My agent once behaved the same way. The problem in my case was a wrong ListenIP parameter in the config file. The IP was not existend on this computer and the agent crashed without any hint in the log files!

[ReefShark]

I set the value ListenIP in the config on several servers where I installed the new agent, but it still crashes...

[ReefShark]

So no one has this problem (or a solution)?