I just wanted to duplicate a renku project, hence I decided to fork it to a new location, with a different repo name. The window said that the process failed, but I can actually find the new one in my list of projects, and everything seems to be fine. Hence, might be something broken? Or should be everything ok? In that case, I just wanted to report this error
Thanks for raising this issue! Can you tell us which RenkuLab instance this was on?
There are a two steps to forking a project in RenkuLab. The first forks it in GitLab, the second runs the CI pipeline to create the images for the interactive sessions. It sounds like the first step succeeded, but the second failed.
I ran into the same problem. Does fork the project, but cant start an environment afterwards. What I also find strange is that it states the Renku Version is 0.12.0 and up to date ? I tried to update to 15.1 manually, not sure if it worked, at least this overview still states 12.0. Furthermore, updating the Templates runs forever, or at least longer than my battery lasts (± 4 hours)…
I copied the dockerfile from another project I have ( and which can be forked), but didnt change the version or solve the problem.
The forked docker image seems not to be available, is there a way I can force to build this image? The status writes:
git error: stderr: ‘fatal: ambiguous argument ‘origin/master’: unknown revision or path not in the working tree. Use ‘–’ to separate paths from revisions, like this: ‘git […] – […]’’
In this case, the easiest way to fix the problem is to push a commit to the repository. One way to do this is to launch an environment using the base image, then modify some file (for example the README), commit and push that.
If you do that, then the pipeline to build the image will be triggered.
Sorry that this is a bit complicated. We will work on making it easier to recover from this error. The instructions above a quite compressed. Let us know if you have any questions, and will fill in any details you need.
Thank you for the fast and clear reply! The workaround worked fine. One more question,
Was this error triggered by a mistake/failure on my side in the first place, respectively, should I look into it deeper, or will it probably be solved by a feature release?
The problem is caused by the infrastructure, not by anything you did. We still do not completely understand why this happens. If we figure it out, we will fix the problem at the source; in the meantime, we plan to make some improvements so users can more easily recover if it does happen.