My project: balena Minecraft Server

Project help
51

Solved it. Easier than I thought. There is a ‘Purge Data’ option on the Balena dashboard. User that and the mc-server generated a new world - just in case anybody runs into same newbie ‘challenge’.

52

OMG thanks! I was so annoyed with this!

53

But it says that it only supports java 8, how do i change java11 to 8? In ubuntu was with the jdk but here how can I do it?

54
55

Hello, the project worked fine for me , but i coudn’t connect directly to the game. It says unknown host. I can’t use ssh from dashboard it gives me ERROR 111 Connection Refused. I made a post on the forum : Need Help with minecraft server running on raspberrypi4 for a more detailed description. My pc is running pop_os and is connected via eth and rpi is connected via wifi.

56

Hi there,
works great on my Pi4 - thaks for that :slight_smile:
One question though: I’d like to setup my Pi to act as an access point so my son an his friends can play together without an extra WiFi-Router. I found this tuto (How-To: Turn a Raspberry Pi into a WiFi router | Raspberry Pi HQ), but I don’t know how to follow the instructions on an balena-managed server/OS…any help will be greatly appreciated.
Greets,
Buzzy

57

@buzzyheb Hi there! The idea is good, but will take some time to figure out. I tried half a year ago to get this stuff working, but gave up after weeks of no success. But luckily the balena team seems to have figured it out!

I think you should be able to set this up easily by merging the docker compose files and folders. Balena made a guide on this too:

58

Hello, I am unable to find ENABLE_UPDATE anywhere. Can anyone tell me where it is? Thank you.

Everything else goes well accept, MC server.

59

Hi Alex,
thanks for your response and sorry for the amout of time passed since then - I simply missed it :expressionless:

One more thing: I’ve been asked to reset the MC-server - is this something to do in the MC-console or via the balena-“container”?

Thanks & Regards,
Buzzy

60

Hi Buzzy, sorry for the delayed reply here! When you say “reset the MC Server”, are you referring to simply rebooting the device? If so, you can just power cycle it. Alternatively, if you mean “delete the world and start over”, then one way would be to re-flash the SD Card, which will format the card and overwrite it, and then ultimately deploy all new containers…thus starting you over.

62

Good morning (and -again- sry for the delay, too),
I had the start-all-overthing in mind… but I was hoping that it was more like “delete files X and Y and folder"Z and here you go” :wink:
Thanks,
Buzzy

63

You can delete all server files using scp, there is a tutorial on how to access scp in the readme file. (there is no such command for it built in). If you can’t figure that out, then just open a terminal in the mc-server container, ctrl + c the rcon client and run the following: cd /usr/src/serverfiles/ and then rm -rf *

Hope it helps!
Alex

64

I’ll give it a try, thanks :slight_smile:
Greets,
Buzzy

65

Hi,

I’ve had this up and running on RPi 4 for a few weeks - thanks! I’d really appreciate some help - this morning, the MC server can’t be connected to, and looking at the dashboard:
image
both the mc-server and scp-server are ‘exited’. I’ve tried restarting the services, and rebooting the device, but neither start, the line 'Starting service ‘mc-server sha256:[etc]’ appears in the log window, but no further info. Any idea how I go about getting things working again?

I can access the RPi through SSH, but cannot find log files that might help me. I’m completely new to the world of containers, so don’t know how to get started.

66

OK, logged in via SSH, ran df -h:

Filesystem                      Size  Used Avail Use% Mounted on
devtmpfs                        817M     0  817M   0% /dev
tmpfs                           951M  4.0K  951M   1% /tmp
/dev/mmcblk0p2                  300M  286M     0 100% /mnt/sysroot/active
/dev/disk/by-state/resin-state   19M  395K   17M   3% /mnt/state
overlay                         300M  286M     0 100% /
/dev/mmcblk0p6                   28G  1.6G   25G   7% /mnt/data
tmpfs                           951M     0  951M   0% /dev/shm
tmpfs                           951M  9.2M  942M   1% /run
tmpfs                           951M     0  951M   0% /sys/fs/cgroup
/dev/mmcblk0p1                   40M  7.0M   33M  18% /mnt/boot
tmpfs                           951M   24K  951M   1% /var/volatile
/dev/mmcblk0p3                  300M  2.1M  278M   1% /mnt/sysroot/inactive

and it looks like a couple of the mounts are 100% full.

Anyone got any ideas about what might have filled it up (logs? cache? tmp files?)?

These are the top 5 directories by size in /mnt/sysroot/active, but I don’t know what to do with this information!

# du -a /mnt/sysroot/active/ | sort -n -r | head -n 5
290536  /mnt/sysroot/active/
290509  /mnt/sysroot/active/balena
267654  /mnt/sysroot/active/balena/overlay2
267643  /mnt/sysroot/active/balena/overlay2/593055c6eef95ad7a03154222980cd7984738ad21c9fa770bed29bb0106b4103
267641  /mnt/sysroot/active/balena/overlay2/593055c6eef95ad7a03154222980cd7984738ad21c9fa770bed29bb0106b4103/diff

Thanks in advance.

67

Hello @jim1, it is expected that the root partition is completely full. They are mounted as read-only and they are usually 99-100% full when we ship the image. Your data and state partitions have plenty of free space.

If your services aren’t starting I would recommend looking at the supervisor and engine logs to see what’s going on. Something like this:

journalctl -u balena -a

You can find more steps to review the engine and supervisor logs here: