Finch does not support logging driver "none"

[pruszel]

Describe the bug When I use the logging configuration for my compose service with the driver option set to "none", then I get a fatal error and am unable to start my compose service.

Steps to reproduce Given the following compose service definition inside docker-compose.yaml:

version: "3.7"

services:
    elasticsearch:
        image: docker.elastic.co/elasticsearch/elasticsearch:6.5.1
        environment:
            - cluster.name=docker-cluster
            - bootstrap.memory_lock=true
            - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
        ulimits:
            memlock:
                soft: -1
                hard: -1
            nofile:
                soft: 65536
                hard: 65536
        logging:
            driver: "none"
        volumes:
            - esdata:/usr/share/elasticsearch/data:delegated
        ports:
            - 9201:9200
        networks:
            - dev

networks:
    dev:
volumes:
    esdata:

When I run finch compose up elasticsearch I get the following output:

INFO[0000] Ensuring image docker.elastic.co/elasticsearch/elasticsearch:6.5.1
INFO[0000] Creating container myapp-elasticsearch-1
FATA[0000] unknown logging driver "none": not found
FATA[0000] error while creating container myapp-elasticsearch-1: exit status 1
FATA[0000] exit status 1

I also tried to use the logging configuration with the driver: "none" option for a different service and got the same result.

Expected behavior I expected the elasticsearch service to start successfully.

Screenshots or logs N/A

Additional context finch-support-20231031104304.zip

[austinvazquez]

Hi @peterlopez , thanks for the report.

I have looked into the gap and found this would require an upstream change to nerdctl. I have asked maintainers to consider this use case for https://github.com/containerd/nerdctl/issues/1039.

[coderbirju]

'none' logger feature has been accepted upstream, should be available on nerdctl 2.0. #3633