Monitors randomly report state down when the server is UP. You can also see the same failures for tcp-default monitor.

In trace, we can see that if NS sends SYN from core ID 1 and if ACK received on core ID 0 then core ID 0 sends RESET with code 8212 and 1 second later core ID 1 sends RESET with code 9800 which means RESET due to timeout. We can see same in events.

You can see the following event in newnslog, if we grep for the service:

DOWN; Last response: Failure - Time out during TCP connection establishment stage

Upgrade to 11.1 57.11 or later version to resolve this issue.

Problem Cause

This issue is specific to 8900 platform.