OpenBalena first start error

About a week I want to start openBalena server. I prepared VPS based on docker on which whole server will run. I redirected mydomain.com to VPS and then go through whole setup. Configuration was ok, but when I entered command “./scripts/compose up -d” I get this error.

ERROR: for openbalena_s3_1 Cannot start service s3: OCI runtime create failed: container_linux.go:345: starting container process caused “process_linux.go:430: container init caused “rootfs_linux.go:70: creating device nodes caused \“open /var/lib/docker/vfs/dir/592b5d7bf96c3a83cce2b064b79c46be88cc355a05e33d1a443cb4ce73af72ce/dev/tty: no such device or address\”””: unknown

ERROR: for s3 Cannot start service s3: OCI runtime create failed: container_linux.go:345: starting container process caused “process_linux.go:430: container init caused “rootfs_linux.go:70: creating device nodes caused \“open /var/lib/docker/vfs/dir/592b5d7bf96c3a83cce2b064b79c46be88cc355a05e33d1a443cb4ce73af72ce/dev/tty: no such device or address\”””: unknown

I don’t know where is problem. Maybe it is bug, maybe my bad configuration. I will be very grateful for any advice.

Hey @romciopaluszek, when you say a VPS based on docker, what exactly do you mean?

Do any of the other services start up? I found this issue which may be relevant https://github.com/lxc/lxd/issues/4531#issuecomment-386604120 but if other containers do start up I don’t think it is the same problem.

My mistake. My VPS is based on LXC. Operating system runs in LXC.

All others containers works. Only S3 containers fails.

Hi,

In theory, I believe this should all work, and it looks like the underlying component used in the S3 service (Minio), maybe be attempting to carry out something terminal based which it can’t. I’ve contacted the engineers responsible for OpenBalena, as I know one of them has run this on a VPS in the past. We’ll get back to you as soon as we have any more information.

Best regards, Heds

Hi again,

We’ve had a discussion internally, and we’ve seen these problems before. Unfortunately, we’ve not spent much time looking at running OpenBalena on LXC systems, as Moby generally requires a dedicated kernel to run against. It appears that most other containers are running from what you say though, so what may be worth trying is the latest version of the S3 service (I made some changes a couple of months back to update the version and build reqs for this, and it may make a difference).

You can do this by altering the version of S3 used in OpenBalena by altering the open-balena/compose/versions file before generating your OpenBalena instance. Update the line export OPENBALENA_S3_VERSION_TAG=v2.6.2 with export OPENBALENA_S3_VERSION_TAG=v2.8.0. If this does work for you, we’d be very grateful if you could let us know!

Best regards, Heds

Hi,

Having checked, the S3 service does not actually get used in OpenBalena yet (we are intending to use it for registry storage in the near future), so you should actually just be able to remove it from the services in the open-balena/compose/services.yml file completely, which should allow you to run OpenBalena without it (with no detrimental effect).

However, please note that this may mean that when we do move to future versions using it, an update path may become difficult or unviable.

Best regards, Heds

When I changed versions, problem wasn’t solved. When I removed S3 service I get next error.

openbalena_redis_1 is up-to-date
openbalena_cert-provider_1 is up-to-date
openbalena_db_1 is up-to-date
Creating openbalena_api_1 …
Creating openbalena_registry_1 …

ERROR: for openbalena_api_1 UnixHTTPConnectionPool(host=‘localhost’, port=None): Read timed out. (read timeout=60)

ERROR: for openbalena_registry_1 UnixHTTPConnectionPool(host=‘localhost’, port=None): Read timed out. (read timeout=60)

ERROR: for api UnixHTTPConnectionPool(host=‘localhost’, port=None): Read timed out. (read timeout=60)

ERROR: for registry UnixHTTPConnectionPool(host=‘localhost’, port=None): Read timed out. (read timeout=60)
ERROR: An HTTP request took too long to complete. Retry with --verbose to obtain debug information.
If you encounter this issue regularly because of slow network conditions, consider setting COMPOSE_HTTP_TIMEOUT to a higher value (current value: 60).

Hi, thank you for your findings there. We will take a note and will let you know if things will change from our side