Open fleischsalatinspace opened 2 years ago
@fleischsalatinspace still experiencing this one?
I assume the issue is still there but since creating this issue i have configured an automatic restart of the containers every 24 hours as a workaround :shrug:
Did you do that in the compose file, or outside the project using cron etc?
Version: master @ v2.1.4 Setup: prod
Description
There are several reports of performance issues with route calculation and i can confirm this issue on several of my instances including the public one with different user tiers. The first POST to /api/rest/Route on a populated map with 20 users did peak with 19 seconds. Following POSTs were short with 1 sec to 5 sec on same or different systems. After checking monitoring and logs, the only parameters affecting the timing of the POSTs looks to be: uptime of Pathfinder Number of systems on map (maybe number of routes per system?) (maybe configured [PATHFINDER.ROUTE] parameters, we both run default values of pathfinder-containers)
There are no logs related to the issue and no logs in esi*.log
Not sure if this is a Pathfinder issue or an issue resulting from pathfinder-containers default configuration
Workaround/Fix
Restart Pathfinder. Clearing Redis cache does not resolve the issue