2 similar agents behave diffently

[hensor]

I've followed this tutorial on a Zabbix server and 2 agents:

https://www.digitalocean.com/community/tutorials/how-to-install-zabbix-on-ubuntu-configure-it-to-monitor-multiple-vps-servers

All 3 servers are completely new Ubuntu 14.04 LTS with no firewall/IP-tables rules.

My first agent shows this in the log when restarting:
19268:20141030:122027.549 Got signal [signal:15(SIGTERM),sender_pid:19980,sender_uid:0,r eason:0]. Exiting ...
19268:20141030:122029.550 Zabbix Agent stopped. Zabbix 1.8.11 (revision 26205).
19996:20141030:122029.571 Starting Zabbix Agent [metadata]. Zabbix 1.8.11 (revision 26205).
19998:20141030:122029.571 agent #1 started [active checks]
19997:20141030:122029.572 agent #0 started [collector]

In Zabbix server the status of that agent is error.
Connection refused to the IP.

My second agent shows this in the log when restarting:
1892:20141030:122244.943 Got signal [signal:15(SIGTERM),sender_pid:1,sender_uid:0,reaso n:0]. Exiting ...
1891:20141030:122244.944 Got signal [signal:15(SIGTERM),sender_pid:1,sender_uid:0,reaso n:0]. Exiting ...
1890:20141030:122244.947 Got signal [signal:15(SIGTERM),sender_pid:1,sender_uid:0,reaso n:0]. Exiting ...
1887:20141030:122244.947 One child process died (PID:1891,exitcode/signal:255). Exiting ...
1893:20141030:122244.948 Got signal [signal:15(SIGTERM),sender_pid:1,sender_uid:0,reaso n:0]. Exiting ...
1889:20141030:122244.953 Got signal [signal:15(SIGTERM),sender_pid:1,sender_uid:0,reaso n:0]. Exiting ...
1887:20141030:122244.955 Zabbix Agent stopped. Zabbix 2.2.2 (revision 42525).
1996:20141030:122245.012 Starting Zabbix Agent [3d]. Zabbix 2.2.2 (revision 42525).
1996:20141030:122245.012 using configuration file: /etc/zabbix/zabbix_agentd.conf
1999:20141030:122245.028 agent #1 started[listener #1]
1998:20141030:122245.028 agent #0 started [collector]
2000:20141030:122245.031 agent #2 started[listener #2]
2001:20141030:122245.032 agent #3 started[listener #3]
2002:20141030:122245.032 agent #4 started [active checks #1]
2002:20141030:122245.036 active check configuration update from [127.0.0.1:10051] started to fail (cannot connect to [[127.0.0.1]:10051]: [111] Connection refused)

In Zabbix server the status of that agent is OK.

That seems weird. What to due? I'm new to Zabbix, but would like both servers to be monitored actively.

[gmoynihan]

The Zabbix agent was installed differently on each host. The first host's agent is running Zabbix version 1.8.11, the second host is running Zabbix 2.2.2

Hope this helps,

Greg Moynihan

[hensor]

That's weird as both servers are installed at the same time. However, it is from a cloud service over the internet, so I will have a talk with them. But updating the old to a new version did indeed fix the issue!

Thank you.

[gmoynihan]

Glad to help!

Somehow, those zabbix binaries are ~2 years different, so your hosts were built differently. For an example, each host could have been built with different choices from a PXE install menu. Or, those hosts were cloned different templates. Or, a post-installation process such as puppet was forgotten, or had errors. Or something different, of course. You'll probably want to figure out those details.

The next thing I would do to see how much they differ is run diff on each hosts' /etc/lsb-release files. Also, diff the output of dpkg --list.

[aib]

Or may be different repositories was used to setup Zabbix Agent.
So far there are at list four (4) different version and repositories existing on the same Zabbix repo server.

http://repo.zabbix.com/zabbix/[LIST][*][DIR] 1.8/                    26-Jul-2012 11:47    -   [*][DIR] 2.0/                    16-Jul-2013 18:23    -   [*][DIR] 2.2/                    12-Nov-2013 10:25    -   [*][DIR] 2.4/                    11-Sep-2014 11:17    -[/LIST]