Closed jh1950 closed 7 months ago
sonroyaalmerol
commented
7 months ago I haven't tested this extensively yet but I get better stability with the BOX64_DYNAREC_STRONGMEM env var set to 3. Try adding that to your docker-compose file and let me know if it helps.
jh1950
commented
7 months ago I added it to docker-compose.yml as below, but the same issue still happens :(
environment:
...
- BOX64_DYNAREC_STRONGMEM=3Yup, mine crashed as well after a restart with only that env var. I've been playing around Box64's env vars to find the best combinations for Palworld. I'm almost certain it's some sort of combination between BOX64_DYNAREC_STRONGMEM and BOX64_DYNAREC_BIGBLOCK with BOX64_DYNAREC_BLEEDING_EDGE=0.
Currently, I have my server running with these and has been stable for a while:
environment:
- BOX64_DYNAREC_STRONGMEM=3
- BOX64_DYNAREC_BIGBLOCK=0
- BOX64_DYNAREC_BLEEDING_EDGE=0Try out these settings. If you could help me test different combinations as well, that would be great. Box64 has documentations for the accepted values. I'll set the right configs as defaults as soon as we confirm the right ones.
FYI: One of the main reasons why I opted out of FEX-Emu (which is what TeriyakiGod is using) is because of it needing an x86-64 RootFS to run. That would increase both build time and image size by a huge amount (could reach around 3gb or more). Box64 feels more appropriate in a Docker container.
sonroyaalmerol
commented
7 months ago Another thing you could try is to free up your host machine's memory manually before running the container.
# Make sure the container has been stopped
docker compose down
# Clear PageCache, dentries and inodes
sudo sh -c 'echo 3 > /proc/sys/vm/drop_caches'Yup, mine crashed as well after a restart with only that env var. I've been playing around Box64's env vars to find the best combinations for Palworld. I'm almost certain it's some sort of combination between
BOX64_DYNAREC_STRONGMEMandBOX64_DYNAREC_BIGBLOCKwithBOX64_DYNAREC_BLEEDING_EDGE=0.Currently, I have my server running with these and has been stable for a while:
environment: - BOX64_DYNAREC_STRONGMEM=3 - BOX64_DYNAREC_BIGBLOCK=0 - BOX64_DYNAREC_BLEEDING_EDGE=0Try out these settings. If you could help me test different combinations as well, that would be great. Box64 has documentations for the accepted values. I'll set the right configs as defaults as soon as we confirm the right ones.
FYI: One of the main reasons why I opted out of FEX-Emu (which is what TeriyakiGod is using) is because of it needing an x86-64 RootFS to run. That would increase both build time and image size by a huge amount (could reach around 3gb or more). Box64 feels more appropriate in a Docker container.
Yes, if I use FEX-Emu, it takes a long time for groupmod and chown commands because of RootFS, so I'm deleting this part and using it. So I tried this image and it goes very fast!
This alone seems to be working fine. I used to have problems with every 2-3 reboots, but now I have no problem with 10 reboots!!
I'm glad the issue has been resolved! Thank you :)
Describe the bug
I saw that there was an ARM64 version of Docker file and tried it. This works fine, but sometimes issues appear when restarting.
To Reproduce
Steps to reproduce the behavior:
docker compose up -dOS information
Hardware information
Oracle Cloud (VM.Standard.A1.Flex)
docker-compose.yml contents
Container/Host Logs
Additional context
Sometimes it only occurs up to
Malloc Size=262146 LargeMemoryPoolOffset=262162, and in this case, a timeout occurs when accessing. There is no point restarting the cloud server, and the numbers such as262146do not change.Until now, I ran a server with a proper combination of this repository and Dockerfile files from the https://github.com/TeriyakiGod/steamcmd-docker-arm64 repository, and this issue has never occurred. So it doesn't seem to be my cloud server problem.