- repro builds to use fixed uid=1000 inside the container
- in case the file permissions leak into the binaries, they are still reproducible
- chown 1000:1000 fresh_clone
- repro builds to create fresh_clone dir outside git clone
- otherwise the local dev build would still interact with the fresh_clone dir
- due to e.g. recursive "find -exec touch",
- and even the "docker build" cmd itself would try to stat/read it
- see https://github.com/docker/for-linux/issues/380
- and "rm -rf fresh_clone" needs sudo if the host uid is not 1000
- this way the local dev build does not need sudo
to recap:
- local dev builds use the host userid inside the container, directly operate on the project dir
- does not need sudo
- repro builds create a fresh git clone, chown it to 1000, and use userid=1000 inside the container
- if the host userid is 1000, does not need sudo
- otherwise, needs sudo
closes https://github.com/spesmilo/electrum/issues/8261
reproducibility probably needs a hardcoded userid
Also, move the UID arg later in the dockerfiles, for better caching.
(if local dev build and repro build set different UIDs, the build caches
will diverge at that step)
This includes two logically separate changes:
- on the host, try not to require sudo when running the build scripts
- namely when interacting with the docker daemon, this requires
the unix user on the host to be part of the `docker` group
- this solves part of https://github.com/spesmilo/electrum/issues/7602
- while running inside the docker containers, do not run as root
- this means that e.g. files created in mounted folders should
no longer be owned by root on the host
- there is some code duplication involved here - not sure
how it could be deduped.