I'm really happy GitLab exists, and that they do so many things in the open. But equally, this is an example of something I see surprisingly often: incredibly long-running feature discussions.
In this case, a scheduled pipeline can fail with no visible error, including on the page where you can trigger the pipeline manually. This seems a pretty bad situation, as there's no way of knowing it failed without looking for it in the pipelines list.
It appears from the discussion that they've decided to send an email. It's taken three years to get to this point.
I'd love to know from anyone close to the development - is this normal? Is it as frustrating as it looks, or is it just how it has to be?
1 comment
[ 1.9 ms ] story [ 15.9 ms ] threadIn this case, a scheduled pipeline can fail with no visible error, including on the page where you can trigger the pipeline manually. This seems a pretty bad situation, as there's no way of knowing it failed without looking for it in the pipelines list.
It appears from the discussion that they've decided to send an email. It's taken three years to get to this point.
I'd love to know from anyone close to the development - is this normal? Is it as frustrating as it looks, or is it just how it has to be?