I was holding off on posting a bug report on this, because we were led to believe our issue was due to network connectivity issues on our server provider's side. However, after we witnessed 2 very similar cases (VEC Nurburg and FFSCA Daytona), I decided to post this after all in the hopes this will be looked into and fixed for future endurance events. Two weeks ago, we hosted (or tried to host) the GEM Nürburgring 24H. It ran well for about 8 hours, until it started to get dark. Around 9 PM CEST, we suffered a mass disconnect - which also affected our TeamSpeak, hosted on the same server - while the dedi continued to run without apparent issues, so we red flagged the race. Believing the problem to be a temporary network connection loss, we performed a reboot and restarted the race. To our astonishment, the issue repeated itself about 15 minutes after the restart, so we red flagged the race again. Still believing this was caused by network issues on our server, we switched to our backup server - located in a different data center - and restarted once again, but in vain. About 5 minutes into the race, the server crashed again and we decided to terminate the race. Technical specs (both servers): - VPS, server shared with 1 other VM - AMD EPYC 7282 (6 cores) at 2.8 GHz - 16 GB DDR4-3000 - 200 GB NVMe storage - 1 Gbps, unlimited bandwidth Active Plugins: - LiveRacers - Weather Plugin - TrackIR (native plugin) Unfortunately, we don't have trace logs of these sessions. Attached are the result logs from the 2 restarts, the first one has not been created due to the server crashing after the mass disconnect.
After further discussion with one of the hosts from FFSCA and some of the drivers that participated in the GEM N24, we found another similarity between both races: after people were kicked by the server and rejoined, their brake bias and engine mix were suddenly locked (not changeable onboard).
It's been a month since I've posted this and many other event hosts have experienced the same (or very similar) problems. Is there any update on the situation and when we can expect this issue to be fixed? I hoped it would be with the Q2 update, but there's nothing about this in the changelog.
I don't think we have any way of logging connection issues, but I have seen an increased number of reports. I've been working with some people to try to fix their issues, but some seem to have all the correct settings, so it's a head scratcher for sure.
