I am having a problem lately with DB connection issues. I tried looking for cases related to "QSConnectionLostConnection" but this seems to be undocumented.
There's 2 scenarios where this issue occurs. I'm not sure if they're related but they share some commonalities:
- They seem to only occurr on the secondary server host, "alttrans02".
- They occur at different points of the translation.
- The problem is gone when re-running the translation.
1. Reading from DB: Please refer to job_20627 and job_21289 logs.
These end in failure but under FME Server console, it's still "running" and I have to manually cancel them. If the job is still "running" and I rerun the translation, the second job also fails at the same spot in the translation and ends up in the same "running" status.
I only have server logs for the second job - please see ALTTRANS01_fmeserver_2018-09-27_064438.
2. Writing to DB: Please refer to job_20836 log.
Environment
- FME Workbench and FME Server 2016.1.3.2 #16717 win64
- Distributed FME Server: 2 engines x 2 hosts
- PostGIS/PostgreSQL 9.6.1, compiled by Visual C++ build 1800, 64-bit
- using Named DB connection