You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Roundtrip times for running tests on Travis are really annoying - here is a random PR #311 as an example - with Travis tests taking around one hour, while Circle settling down in under 10 minutes.
This significantly lowers productivity by sometimes just waiting and a substantial organizational extra effort if one just wants to review and or rebase 3-4 PRs on a same day.
Furthermore it also significantly discourages going on on more roundtrips to make improvements on current work, which at the end contributes to lower overall code quality.
Should we consider dropping Travis tests and just rely on Circle or does this bring significant extra benefits I am maybe not seeing here?
The text was updated successfully, but these errors were encountered:
There is no reason for travis if all the same cases are covered by circleci. The only benefit could be keep it disabled so it can be enabled in case of a circleci outage, however that would mean the configuration must be kept up to date and for that exceptional case I do not think it is worth the effort.
Ok, I've the set the mandatory tests to pass for master branch merges from Travis to Circle and deactivated the Travis builds directly on the Travis settings page (so not touching the integration on GitHub).
Roundtrip times for running tests on Travis are really annoying - here is a random PR #311 as an example - with Travis tests taking around one hour, while Circle settling down in under 10 minutes.
This significantly lowers productivity by sometimes just waiting and a substantial organizational extra effort if one just wants to review and or rebase 3-4 PRs on a same day.
Furthermore it also significantly discourages going on on more roundtrips to make improvements on current work, which at the end contributes to lower overall code quality.
Should we consider dropping Travis tests and just rely on Circle or does this bring significant extra benefits I am maybe not seeing here?
The text was updated successfully, but these errors were encountered: