Running command in docker container results in directly shutting down the docker container

Question

[stan.aanhane]

Im currently struggling to run a container and execute a command within this container.I'm trying to use NanoSNP (https://github.com/huangnengCSU/NanoSNP), which is an image available from dockerhub. After fulling this docker, the following command would be sufficient to run the the container:

docker run \
-v "${INPUT_DIR}":"${INPUT_DIR}" \
-v "${OUTPUT_DIR}":"${OUTPUT_DIR}" \
--gpus all \ huangnengcsu/nanosnp:v2.1-gpu \ run_caller.sh \
-b "${INPUT_DIR}/input.bam" \
-f "${INPUT_DIR}/reference.fa" \
-t "${THREADS}" \
-c "${COVERAGE}" \
-o "${OUTPUT_DIR}"

What i've been using to start the docker is the following:

docker run snp \
huangnengcsu/nanosnp:v2.1-gpu \
    run_caller.sh \
        --bam_fn=/home/stanaanhane/anaconda3/envs/SNPcalling/Code_SNP/BRCA1_MAP.sorted.bam \
            --ref_fn=/home/stanaanhane/anaconda3/envs/SNPcalling/Code_SNP/BRCA1.fasta \
                --output=/home/stanaanhane/anaconda3/envs/SNPcalling/Data \
                    --threads==10 \
                        --coverage==1

But when started, I see the docker container starting, but almost directly stop. The docker ps -a looks like this:

99cd4e272f19   huangnengcsu/nanosnp:v2.1-gpu   "/bin/bash"   19 hours ago   Exited (0) 18 hours ago               snp
446add2244e4   ubuntu                          "bash"        19 hours ago   Exited (137) 18 hours ago             festive_kirch
013d00f65681   hello-world                     "/hello"      23 hours ago   Exited (0) 23 hours ago               upbeat_cannon
78481d30f8a0   hello-world                     "/hello"      23 hours ago   Exited (0) 23 hours ago               loving_mclean
1aba1e8fa496   hello-world                     "/hello"      23 hours ago   Exited (0) 23 hours ago               admiring_visvesvaraya

I've tried looking around, but can't find the solution. If anyone has a suggestion, please let me know.

Thanks!

[GenoMax]

Looks like this requires a GPU to work. Do you have one available on your machine and is it configured for use with docker?

[stan.aanhane]

GenoMax I do have an RTX 3050 ti (notebook). But when adding the -- gpus all command above huangnengcsu/nanosnp:v2.1-gpu \, it shows the following error code:

docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]]. ERRO[0000] error waiting for container: context canceled

I tried installing the linux GPU driver, but bash shows its not compatible. Im using WSL btw.

[GenoMax]

This is pretty tricky (trying to use GPU with docker under WSL). I am not a docker expert but it sounds like you will need to do some configuration on docker side to use this

https://stackoverflow.com/questions/70761192/docker-compose-equivalent-of-docker-run-gpu-all-option

[stan.aanhane]

It seems like the problem is either one of the two following problems:

1. I cant use sudo apt-get install nvidia-docker2, it will give me the following error: Unable to locate package nvidia-docker2

The solution (with the needed repository) has migrated on the Nvidia website, so I haven't been able to locate that.

1. When running nvidia -smi it shows the following:

   +-----------------------------------------------------------------------------+ | NVIDIA-SMI 525.85.05 Driver Version: 528.24 CUDA Version: 12.0 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA GeForce ... On | 00000000:01:00.0 Off | N/A | | N/A 49C P8 10W / 45W | 220MiB / 4096MiB | 5% Default | | | | N/A | +-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 27 G /Xwayland N/A | +-----------------------------------------------------------------------------+

This suggests that the GPU is recognised, but when running the benchmark

sudo docker run --rm --gpus all nvidia/cuda:11.6.2-base-ubuntu20.04 nvidia-smi

it shows the following error:

docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].

so its either that my GPU is not detected or not compatible (RTX 3050 ti notebook).