I've read the other 2 posts on this error and we have very similar issues with some users having "performance" issues with vanishing zombies, players phasing in and out, etc. I understand this is the server growing the buffer for new/more information but how do we prevent this from happening? We were paying for hosting and never seen this issue, since we switch to our own dedicated server, we are now experiencing this. This dedicated server is on an i7 13700 with 128GB of memory on Windows Server 2022. Is there a way to tell Java to utilize more memory so it doesn't need to do this?
#IsoChunk.Save BufferOverflowException
35 messages · Page 1 of 1 (latest)
How do we make the server not "need" to grow this buffer? There must be a Java startup parameter to resolve this?
You can get the server using more memory by setting the -Xms -Xmx parameters in the .bat (StartServer64.bat (or 32)) file used to start the server, if that's what you're asking. I don't think that would fix a BufferOverflowException but you could give it a shot.
As mentioned above, the .bat file have the -Xmx parameter, which controls the maximum RAM
In our Java start paramaters I have it set to 32768, I could up that some more I guess but, currently, we are only at 20GB used
I am not exactly sure what this UI is.. are you using some kind off hosting service?
I would probably recommend reporting the issue to them if this issue causes problems for you
This is AMP by Cubecoders
The answer would likely be the same in that case depending on the issue you get with the service, as using any sort of game service or manager would usually mean that the installation is not standard
Would an error like this be related?
Possibly, though it is not clear what would be triggering it
I do not often see these specific errors in any server situation
How is your server installed exactly?
Installed using steamcmd as AMP is just a control system. Give you a GUI for settings, etc same as most other hosting solutions
This is from someone on the Cubecoders AMP Discord
Hm, corruptions really should not happen often
Unless, do you have any mods enabled?
Oh boy, do we ever...
Are you able to perhaps replicate this issue in that case? Does anything that you do specifically trigger it?
The game saving almost always triggers it. I see 3 of those error messages then the save happens
Are you able to install a new server on the server machine in that case? Just to use as a test server. You can use a different name / port so it runs alongside this original one
Try loading it without any mods and then run the save command to see if it happens
Will do, I have a feeling I know how this will go heh, I'm fearing we have corruption
So we were having some weird performance issues. Upon reading a post about the mod Advanced Trajectory, we disabled the mod this AM and the errors went away... Coincidence? I don't know but performance is better and errors are gone
A lot of mods may cause different performance issues. I would not really be able to help diagnose those since generally there would be no log or anything pointing to the mods
Experimentation is the only real way to find it out
Ugh... the errors are back. Very odd, must have been a coincidence that it stopped giving errors in-between the restarts. So it is back to getting 1 error every save and now we are getting a "port 16262 is closed" which nothing has changed and we didn't have that issue before
Can you make sure that your players are not using the "Use Steam Relay" option when connecting to the server?
As that can cause the error message
Otherwise it would just be caused by the port being closed
Alright, I will let people know. Also, I can type "save" in the console 10 times and every time we get a single IsoChunk.Save: BufferOverflowException, growing ByteBuffer. Any idea where I should start looking to find what is causing that issue?
Interesting... after a restart we do NOT get that error when I try to manually save hmm
Hm, its as if it happens once a specific area is loaded and trying to be saved perhaps?
Yeah, I'm working on manually resetting some areas. See if I can track it down the cause of this. Wish there was some script to do database repairs/report a problematic area 😉