Improved sync speed by resolving the N+1 query issues.
Solves #1402 and Solves #1453
With this change there is just one query done to retreive all the
important data, and matching is done in-code/memory.
With a very large database the sync time went down about 3 times.
Also updated misc crates and Github Actions versions.
- Changed nightly to stable in Dockerfile and Workflow
- Updated Dockerfile to use stable and updated ENV's
- Removed 0.0.0.0 as default addr it now uses ROCKET_ADDRESS or the default
- Updated Github Workflow actions to the latest versions
- Updated Hadolint version
- Re-orderd the Cargo.toml file a bit and put libs together which are linked
- Updated some libs
- Updated .dockerignore file
- Split Debian and Alpine into different build matrix
This starts building both Debian and Alpine based images at the same time
- Make use of Docker BuildKit, which improves speed also.
- Use BuildKit caching for Rust Cargo across docker images.
This prevents downloading the same crates multiple times.
- Use Github Actions Services to start a docker registry, starting it
via the build script sometimes caused issues.
- Updated the Build workflow to use Ubuntu 20.04 which is more close to
the Bullseye Debian release regarding package versions.
Since docker hub stopped Autobuild, we need to switch to something else.
This will trigger building of images on Github Actions and pushes them
to Docker Hub.
You only need to add 3 secrets before you merge this PR to have it working directly.
- DOCKERHUB_USERNAME : The username of the account you are going to push the builds to
- DOCKERHUB_TOKEN : The token needed to login and push builds
- DOCKERHUB_REPO : The repo name in the following form `index.docker.io/<user>/<repo>`
So for vaultwarden that would be `index.docker.io/vaultwarden/server`
Also some small modifications to the other workflows.
