Join GitHub today
GitHub is home to over 40 million developers working together to host and review code, manage projects, and build software together.Sign up
Fix windows CI & lengthen verdaccio timeouts #3421
Some CI tuning for the "allowed failure" jobs that stems from Ryan's work on Github Actions.
Type of change
ryanio left a comment
Did you want to add a few more zeroes to keepAliveMsecs?
Did you by any chance try if other values like max_timeout, max_fail, or server: keepAliveTimeout would additionally help at all? https://github.com/ryanio/web3.js/blob/c9e9c3051e88f0ad07a81d9f6c78fcb488a19b85/verdaccio.yml
@ryanio Ah thanks. Looking at the logs you ran, my impression was that the other timeouts (e.g 10m) weren't getting hit. Is anything waiting 10 minutes in those jobs?
Tbh I don't really have any idea what's going on
What is your opinion about leaving the allowed failures in Travis and using GH Actions for the real jobs? Only reason I suggest this is that Travis' UI seems slightly better for that use-case - you can easily see if the job failed but the run as a whole passes.
In my runs it would randomly time out on me when downloading deps, so I bumped everything to 10mins to play it safe, but I never saw it wait that long. It would usually just immediately fail on retries up to the
Sure that seems nice, however unfortunately last I heard the
Oh ok. I've ported over a couple improvements you made to