Summary:
I started to use the new MongoDB plugin.
Everything works fine for several hours, then after a while I get only "ZBX_NOTSUPPORTED: timeout occurred".
I have to restart the zabbix agent to get it working again.

My environment:

• Zabbix Agent 2 version: 6.2.0
• MongoDB plugin version: 1.0.0 (branch: master, commit: 21d652197aba9a54f4af45424a8d2ef98f5539b4)

(Faster) Reproduction:
It is possible to reproduce the bug in a few seconds/minutes by spamming the 'mongodb.server.status' requests:

while [ 1 ]; do
  zabbix_get -s 127.0.0.1 -p 10050 -k 'mongodb.server.status["tcp://localhost:27017","XXX","XXX"]' || break;
done

Details:
Once the bug occurs once, it occurs for all subsequent requests (even when testing the URI of another mongodb server).

$ zabbix_get -s 127.0.0.1 -p 10050 -k 'mongodb.server.status["tcp://localhost:27017","XXX","XXX"]'
ZBX_NOTSUPPORTED: timeout occurred
$ zabbix_get -s 127.0.0.1 -p 10050 -k 'mongodb.server.status["tcp://another.working.mongodb:27017","YYY","YYY"]'
ZBX_NOTSUPPORTED: timeout occurred
$ zabbix_get -s 127.0.0.1 -p 10050 -k 'agent.hostname'
my-hostname 

Note: this only concerns the metrics of MongoDB plugin, the others continue to work.
Also these metrics work fine if I try them at the same time on another zabbix agent that hasn't encountered the bug yet.

Problem, when I set "DebugLevel=5" I can't reproduce the bug anymore (maybe it's a concurrency bug and logging takes enough time to reduce the probability of it occurring?).
And I can reproduce again immediately when I reset to "DebugLevel=3".