remote docker daemon with self-hosted runners? #2242
Unanswered
brianjmurrell
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I have been experimenting with self-hosted runners.
To leverage a global docker cache it seemed to make the most sense to have a single docker daemon running on a (big enough) host and then have all of my runners use
DOCKER_HOST=$docker_daemon_host
to do their docker operations on the single big docker daemon. Docker cache resuse would be maximized that way vs. having a docker daemon and cache on each runner node.But it seems it's impractical due to GHA's use of docker bind mounts (i.e.
docker run ... -v
) when running docker containers. I.e.:and in particular the bind mounts from the above:
as those bind-mount destination paths don't have the same content as they have on the GHA runner since they are bind mounted from the docker daemon host, not the GHA runner.
It seems it might be worthwhile for the runner to issue a warning at startup if it detects that remote docker daemon options (i.e.
DOCKER_HOST=$hostname
as on example) are being used.Thoughts? Is my analysis above correct or am I missing something?
Beta Was this translation helpful? Give feedback.
All reactions