Active Agent Checks Not Working

[Huckleberry]

I'm trying to get the Agent Active checks working on a new CentOS 7 client.
The Agent Passive checks work fine.

These aren't the actual IP addresses, however the real ones are all on the same local network.
Only firewalls involved are firewalld on CentOS, of which I've opened all necessary ports (10051/tcp on the Zabbix Server and 10050/tcp on the Zabbix Agent Server).

Zabbix Server Version: 5.0.0alpha4
Zabbis Agent Version: 5.0.8 (revision d3c78f993a)

I've changed just these lines in the zabbix_agentd.conf file;
Server=192.168.1.1
ServerActive=192.168.1.1
Hostname=NS-Syslog

On the Zabbix server I've been sure to create the host with case sensitive name NS-Syslog.

Here's what the agent log is reporting;

  1691:20210128:161753.281 Starting Zabbix Agent [NS-Syslog]. Zabbix 5.0.8 (revision d3c78f993a).
  1691:20210128:161753.281 **** Enabled features ****
  1691:20210128:161753.281 IPv6 support:          YES
  1691:20210128:161753.281 TLS support:           YES
  1691:20210128:161753.281 **************************
  1691:20210128:161753.281 using configuration file: /etc/zabbix/zabbix_agentd.conf
  1691:20210128:161753.282 agent #0 started [main process]
  1696:20210128:161753.283 agent #5 started [active checks #1]
  1693:20210128:161753.285 agent #2 started[listener #1]
  1694:20210128:161753.286 agent #3 started[listener #2]
  1695:20210128:161753.288 agent #4 started[listener #3]
  1692:20210128:161753.289 agent #1 started [collector]
  1696:20210128:161756.285 active check configuration update from [192.168.1.1:10051] started to fail (ZBX_TCP_READ() timed out)

Using zabbix_sender also seems to fail;

[root@ns-syslog ~]# zabbix_sender -c /etc/zabbix/zabbix_agentd.conf -k trap -o "test" -vv
zabbix_sender [1707]: DEBUG: answer [{"response":"success","info":"processed: 0; failed: 1; total: 1; seconds spent: 0.000131"}]
Response from "192.168.1.1:10051": "processed: 0; failed: 1; total: 1; seconds spent: 0.000131"
sent: 1; skipped: 0; total: 1

I'm able to telnet to 10051 from the Agent to the Server as well;

[root@ns-syslog ~]# telnet 192.168.1.1 10051
Trying 192.168.1.1...
Connected to 192.168.1.1.
Escape character is '^]'.

Connection closed by foreign host.
[root@ns-syslog ~]# telnet 192.168.1.1 10052
Trying 192.168.1.1...
telnet: connect to address 192.168.1.1: No route to host

I can also prove 10050 is opened on the Agent with zabbix_get from the Zabbix Server;

[root@ns-zabbix ~]# zabbix_get -s 192.168.1.2 -k agent.hostname
NS-Syslog

Finally I have SeLinux in Permissive mode on both the Zabbix Server and the Agent I'm attempting to setup the Active checks on.

Can anyone give me their thoughts on why this might not be working?

Thanks,
H

[Huckleberry]

I figured out what I believe was the cause of this.
Having both a source compiled server and a yum package managed server installed.

When I noticed that I exported the database, wiped the server and reinstalled Zabbix via yum only, imported the DB and all working as expected.

Thanks,
H