They say a picture paints a thousand words!
I upgraded from 4.4.7 to 4.4.8 yesterday - A very short time after that, I started getting timeout errors for my AIX servers.
They are up and down - they will be unresponsive for a few minutes, then they will come back again.
I checked the Zabbix logs and saw the standard timeout errors - there have been absolutely no firewall changes and this is only effecting AIX;

The timing is exactly when I upgraded Zabbix to 4.4.8 - ever since then, you can see the availability of the AIX agents is up and down, along with the associated data gathering processes;


We're on AIX 7.1 - the agent we're running on them is 3.4.0. I actually went ahead and upgraded the agent to version 4.2.1 on some of them - it's made no difference.
Side-note - Zabbix seems a little slower when pulling data from more than 24 hours ago. Problably 5 seconds to load the graph, when it used to be instantaneous.
Has anyone else experienced problems?
For additional clarity, as soon as I disable all the AIX servers from monitoring, the processes go back to normal;

Thanks,
Tom
I upgraded from 4.4.7 to 4.4.8 yesterday - A very short time after that, I started getting timeout errors for my AIX servers.
They are up and down - they will be unresponsive for a few minutes, then they will come back again.
I checked the Zabbix logs and saw the standard timeout errors - there have been absolutely no firewall changes and this is only effecting AIX;
The timing is exactly when I upgraded Zabbix to 4.4.8 - ever since then, you can see the availability of the AIX agents is up and down, along with the associated data gathering processes;
We're on AIX 7.1 - the agent we're running on them is 3.4.0. I actually went ahead and upgraded the agent to version 4.2.1 on some of them - it's made no difference.
Side-note - Zabbix seems a little slower when pulling data from more than 24 hours ago. Problably 5 seconds to load the graph, when it used to be instantaneous.
Has anyone else experienced problems?
For additional clarity, as soon as I disable all the AIX servers from monitoring, the processes go back to normal;
Thanks,
Tom