Closed valeriupredoi closed 2 years ago
To be clear, in this case this was a planned maintenance at CEDA. I saw the fails and double-checked this. It did make me wonder: Should we have some kind of status page or log of past CI failures?
good clarification, Klaus - indeed, Jassie eloped that day :grin: And a very good point about the log of past fails too, I'll look into setting that up!
Should we have some kind of status page or log of past CI failures?
It is available here: https://app.circleci.com/insights/github/ESMValGroup/ESMValCore/workflows/nightly/overview?branch=main&reporting-window=last-90-days
The originally reported problem was fixed in #1561, but I suspect that there is a different exception now. I'll make a new pull request to catch that one too.
Should we have some kind of status page or log of past CI failures?
It is available here: https://app.circleci.com/insights/github/ESMValGroup/ESMValCore/workflows/nightly/overview?branch=main&reporting-window=last-90-days
thanks, Bouwe! That is returning a Server Error for me at the mo, but also I'd be looking into a log of the GA tests - those are a bit harder to track after they've failed since one needs to scroll through a lot more test results (since they are 8-10x more numerous)
Maybe you need to sign in on CircleCI to see the test results? You can sign in with your GitHub account.
I did manage to get through - the Server Error is intermittent though - Circle may be having a bad hair day, but those metrics do look good and helpful; I'll see if GA offers something similar, maybe we should add those links somewhere in our docs?
maybe we should add those links somewhere in our docs?
Pull requests improving the docs are always very welcome ;-)
so I've set this up https://meercode.io with ESMValCore's Github Actions as Integration (it has many more CI integrations available) - it's pretty nifty - see attached example screenshot - but this is set up as a user so I am not sure how we can integrate it inside our organization. Just a pointer if you guys want to have a look at it
See test fails - all of them and their dog have died when CEDA archive was down a couple days ago - I thought we fixed this behaviour and got the rerouting done if one ESGF archive was down, I might be slightly out the loop if this was done, but clearly is still causing total test failures. @bouweandela what you reckon? :beer: