EU siege servers constantly crashing

Users who are viewing this thread

The EU siege servers are crashing left and right.
  • crashes frequently occur midgame / - exe stops need relogin
  • transitioning to another round almost every time results in a crash to server selection screen
I've seen some threads mentioning this but they werent specific about the location / game type.
 
Last edited:
don't worry, it has only been an issue for about a year. I'm sure taleworlds sees this as a priority and has their best people on it!
Well it had been in somewhat better shape before :smile:. for example i didnt see the eu server come back up for quite a while today, however it restarted after a while earlier.
as a dev / automation guy I would be glad to put in some free/volunteer work for them if that would mean we could stabilise things a bit. (yeah I know naive even to think about it :sad: )..
 
Well it had been in somewhat better shape before :smile:. for example i didnt see the eu server come back up for quite a while today, however it restarted after a while earlier.
as a dev / automation guy I would be glad to put in some free/volunteer work for them if that would mean we could stabilise things a bit. (yeah I know naive even to think about it :sad: )..
It really seems like some super super poorly coded servers, I imagine fixing it would require a huge refactor. If it were easy I'd hope they would've done it by now, not that it being difficult is a good excuse to just never touch it.
 
super super poorly coded servers
not really. if the code is perfect syntactically there might be unforseen states / unsynced services etc. or simple resource based problems might arise: running out of memory / etc.
app/service/platform logs mostly make it easy to track these kind of issues down. analyitcs and profiling help with the more complicated crashes (if they collect those anyway :smile: )
 
not really. if the code is perfect syntactically there might be unforseen states / unsynced services etc. or simple resource based problems might arise: running out of memory / etc.
app/service/platform logs mostly make it easy to track these kind of issues down. analyitcs and profiling help with the more complicated crashes (if they collect those anyway :smile: )
What do you think 'poorly coded' means? You went on to describe bad code, with the one exception being inadequate hardware.
 
well it can be for example also badly designed / architected which is a hard case in itself. "good code" can lead to disasterous outcomes even then...
'bad code' has all kinds of other indicators:
- eg poorly structured -> hinders readability and maintainability; can be correct otherwise etc etc.
- without clear patterns / conventions
- without language specific "sugar" ~ tends to happen with ppl jumping between different languages on the same day. eg ppl coding in go, c#, c++ and some scripts like python. that can easily lead to some dangerously beahving code at the end of the day. (using features which behave in a very very different manner depending on the env.)
I'm dealing with "old code" (10y or older sometimes) on a daily basis luckily more good then bad.
 
@OurGloriousLeader
from today's changelog:
Crashes

  • Fixed a crash that occurred on round start if a castle gate was destroyed during the warmup phase.
  • Fixed a crash that occurred when entering the team selection screen.
  • Fixed a crash that occurred after loading into Skirmish matches.
So they are actually working on it but it can be a pain in the *everywhere* to nail such bugs down :smile:.
 
Back
Top Bottom