I am posting this message to see if any other users have experienced anything similar to this, and to see if anyone has already narrowed this down further so I can focus my efforts in the right area. Currently, I am watchdogging the indicators to see if they are still getting data on all of their series, but that is obviously suboptimal, and is not really a solution, just a way to be notified that it has happened.
I do not know, when this happens, whether F5 on the chart, or reload all historical data on the chart, or disconnect/reconnect is sufficient to restart it because I've always restarted the platform when this happens, but in time I will try to narrow this down. I am thinking I might have to cause these disconnections by unplugging the ethernet cord or some similar measure in order to create a reproducible test case because when it happens "in the wild" I do not often have time at that moment to do a deep dive on the issue and have to focus on getting everything running again ASAP and that means restarting the platform.
Upon restart, assuming it does not disconnect again during the startup process, and assuming it does not get a pacing violation ("more than 10 API calls of type 'Heartbeat' in 150 seconds"), each of which happen sometimes, everything always restarts fine. If either of those happen, I have learned to simply close NT8 completely and start again.
Is this issue familiar to anyone? Were you able to isolate it any further than I have just described?
Comment