Core (sigsegv) on startup with QNAP and 2.3 series

[riversedge]

On a QNAP using Container Station and both latest and armhf branches will not start. You can watch the docker container launch, but appears to immediately sigsegv and return ExitCode 139. Nothing appears on the screen or logs that's of much use from what I can tell either.

I have tested several other branches that work fine. 1.7.5, 2.2.12, and 2.2.15 will all run. 2.3.35 and 2.3.57 crash immediately. Given I don't even see a bootloader or other text appear, possibly something with Alpine Linux 3.17 bootloader/settings?

docker events just shows:

container start [uuid] (image=nico640/docker-unms:latest, name=docker-unms-1)
container die [uuid] (exitCode=139, image=nico640/docker-unms:latest, name=docker-unms-1)

CPU Info: Annapurna Labs Alpine AL314 Quad-core ARM Cortex-A15 CPU @ 1.70GHz
uname -m returns armv7l

docker -v
Docker version 20.10.22-qnap7, build 57ed8b8

docker inspect docker-unms-1           
[
    {
        "Id": "[UUID]",
        "Created": "2024-01-30T02:23:25.39556586Z",
        "Path": "/init",
        "Args": [],
        "State": {
            "Status": "exited",
            "Running": false,
            "Paused": false,
            "Restarting": false,
            "OOMKilled": false,
            "Dead": false,
            "Pid": 0,
            "ExitCode": 139,
            "Error": "",
            "StartedAt": "2024-01-30T02:23:30.0161311Z",
            "FinishedAt": "2024-01-30T02:23:30.1245157Z"
        },
...

[Nico640]

Hi, sorry for the late reply, do I understand correctly that docker logs docker-unms-1 returns nothing at all? Can you post the output if it does return something?

[riversedge]

Right.   Logs don’t output anything.  I think because it crashes right away when booting.   Perhaps there’s a feature or something in the kernel that the CPU doesn’t like?Thanks,-- WesOn Feb 18, 2024, at 3:09 PM, Nico640 ***@***.***> wrote: Hi, sorry for the late reply, do I understand correctly that docker logs docker-unms-1 returns nothing at all? Can you post the output if it does return something? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>

[Nico640]

Hmm, possible. The only thing I can think of that will cause the whole container to crash would be if the s6-supervise process /init died.

In that case it would be good to know if it also happens when trying to run the base image (nico640/s6-alpine-node), because that would narrow it down quite a bit.

Something like this:

docker run -d --name s6-test-latest nico640/s6-alpine-node:latest

then check if the container crashes or stays running and if there is anything in the logs (there should only be a few lines because the image doesn't actually do anything)

If it does crash, I have pushed a new image on the testing tag with a updated version of s6-supervise, check if that one also crashes:

docker run -d --name s6-test-testing nico640/s6-alpine-node:testing

[riversedge]

The s6-test-latest crashes, immediately. docker container logs s6-test-latest returns nothing at all. Same thing seems to happen with the s6-test-testing one. It downloads, starts, and then just stops with no apparent messages. I you try to start it again, it starts, then just stops - nothing is ever displayed and no logs.