- Jun 04, 2021
-
-
nimrod authored
At this point, it's already processed by GitLab CI so there's little point in running the same thing again.
-
- May 22, 2021
- May 21, 2021
- May 10, 2021
- May 02, 2021
- Apr 29, 2021
-
-
nimrod authored
No need to wait for other jobs to finish.
-
- Apr 28, 2021
-
-
nimrod authored
Until I add Terraform to the image or something.
-
- Apr 24, 2021
- Mar 26, 2021
-
-
nimrod authored
It's done with the pull job and it makes the run job faster.
-
- Mar 23, 2021
-
-
nimrod authored
The end goal is pulling and building in parallel and then deployment is shorter.
-
- Mar 12, 2021
-
-
nimrod authored
- Feb 26, 2021
-
-
nimrod authored
Specifically for GitLab, which takes a long time not only to start but finish its setup process and until it does it returns a 502. Eventually it does finish and is healthy but the job already failed.
-
- Feb 12, 2021
-
-
nimrod authored
The job is now hidden, so you can base on it and set a proper script to run the CLI.
-
- Feb 07, 2021
-
-
nimrod authored
-
- Jan 24, 2021
-
-
nimrod authored
When building ci-images, images on the host are sometimes (in the case of the docker image, always) pulled from the registry (overwriting the tag) but before pushing. So the image pushed is that pulled just now. Instead save the build hash (and pretty much everthing else) and re-tag before pushing (using the hash so it's the right image). While at it, unify the Docker push jobs (the Docker Hub and the personal registry one).
-
nimrod authored
-
nimrod authored
-
nimrod authored
-
- Jan 22, 2021
- Jan 21, 2021
-
-
nimrod authored
-
- Jan 13, 2021
- Jan 07, 2021