Hi,
I have Zabbix setup to monitor 10 Tomcat instances (2 servers with 5 instances running on ports 8080, 8180, 8280, 8380, 8480) using Web Monitoring. It checks each one every 60 seconds.

I have the trigger setup to send notifications if the last 3 checks fail:

I am constantly getting random false positives on these. I ran wireshark for 24 hours capturing all data from the Zabbix server to the Tomcat servers and went back through and verified that I am getting HTTP/1.1 200 OK's and the login screen every single time Zabbix polls the Tomcat instances and they are all under 20ms response times.
The return values for when it fails are 0 for the response code and 0 for the response time. Looking at the graph below, if I go look at the tcpdump for from 08:32 to 08:42, everything looks good.

Note: The images I used are not all of the same instances
Edit: I am using 1.8.3
What am I not doing right?
Thanks for any help.
Steve
I have Zabbix setup to monitor 10 Tomcat instances (2 servers with 5 instances running on ports 8080, 8180, 8280, 8380, 8480) using Web Monitoring. It checks each one every 60 seconds.

I have the trigger setup to send notifications if the last 3 checks fail:

I am constantly getting random false positives on these. I ran wireshark for 24 hours capturing all data from the Zabbix server to the Tomcat servers and went back through and verified that I am getting HTTP/1.1 200 OK's and the login screen every single time Zabbix polls the Tomcat instances and they are all under 20ms response times.
The return values for when it fails are 0 for the response code and 0 for the response time. Looking at the graph below, if I go look at the tcpdump for from 08:32 to 08:42, everything looks good.

Note: The images I used are not all of the same instances
Edit: I am using 1.8.3
What am I not doing right?
Thanks for any help.
Steve
Comment