We have a server with a few people nearing endgame in all-planets and we suddenly started getting desyncs happening out of the blue. We ended up getting to just rubia remaining in the mods list and the desyncs kept happening. No one was on rubia (a couple on vulcanus and one on a ship travelling), but we noticed, though uncertain, that it seemed to happen when one of the people on the server (on vulcanus) entered or exited map view, but not every time, but each time it desynced seemed to be when they changed map view after multiple dozens of tests. No one else could seem to replicate it, just that person.

Unsure what's causing it, and we removed rubia for this game now so we could finish it up since we already did rubia a month ago anyway (ton of fun, loved the planet!), but figured I'd drop this off here so it could be recorded anyway in case others have the same issue.

Thanks for the report.

Are you very certain Rubia is involved here? This sounds extremely strange because the only part of Rubia that could even have the opportunity to cause such a desync has not been altered for ~3 months. I am suspicious that another mod is root cause, but if you can find further information, please let me know.

I would love to diagnose it but considering I have to have the person with the slow internet who does not like to do testing is a bit like pulling teeth... ^.^;

Fairly sure it was rubia as it was one of the only planet mods installed at the time other than just a few QoL ones by the time of the search (binary search indeed up removing almost everything, wanted to continue to try to reduce it but they were fed up enough as it was...).

Considering the oddity that it didn't seem timing based or anyone doing anything but that one person (literally still doing nothing other than that one person testing going into and out of their map and swapping planet surfaces) it feels like some bad state somewhere (might not even be rubia, could even be engine state...). I got one of their desync zips from the start of the testing (which took excessively long as it is due to their internet and it being over 600megs) but I seem to need to update my tooling as the 2.0 binary formats seems to have changed and I haven't gotten time to do that yet.

If I can manage to convince them to do more testing I will give more details though (I'd prefer to reduce it to the minimal mod set and then start pruning code, generally starting with events, to figure out the precise issue before reporting, but most issues don't necessitate needing another player, lol). There was not many mods left to prune though and the remaining ones weren't content mods others than schalls pods and such (which don't have much in the way of code or events other than refilling their ammo, but we had none deployed in the entire save anyway). I really hope I can continue to test with them, desyncs are the most annoying, and usually with a desync report in 1.x I could figure out the mod that caused it most of the time, but it seems 2.x's binary format has changed so I need to figure that out...

At the very least I'm glad it happened after we finished the surface other than biofusion (we just got to that, I landed on rubia myself last weekend to start that processing but left it to figure out a good egg shipping back on monday after building a few buildings, no one was on any modded planet at all when the issue happened, either on vulcanus or a platform/ship, made it easy to strip all the planets out for testing though), the gameplay was wonderfully unique and a lot of fun!

Well, no offense, but I can’t really justify spending time tracking down a desync unless I’m absolutely certain it is a Rubia problem. Given how much multiplayer activity there is with Rubia, I find what you describe to be unlikely to be Rubia’s fault, but you never know. For now, I will simply mark it as unconfirmed unless I can get a definitive report.

Well, no offense, but I can’t really justify spending time tracking down a desync unless I’m absolutely certain it is a Rubia problem. Given how much multiplayer activity there is with Rubia, I find what you describe to be unlikely to be Rubia’s fault, but you never know. For now, I will simply mark it as unconfirmed unless I can get a definitive report.