-
Notifications
You must be signed in to change notification settings - Fork 35
PR builds are failing again... TRAVIS!!!! #517
Comments
Over the past few days we have been seeing errors such as (note this is not consistent on which libraries will fail from what I can see):
This is resulting from the make target: |
Was about to open an issue, but it looks like others beat me to it (~a day ago) at travis-ci/travis-ci#6641. Looks like we are not the only ones, so it's not on our end. |
Link to sbt/sbt#2758 as well |
The sbt build started working again this morning. Something to do with HTTPS->HTTP redirects not being allowed, though it's not clear who fixed this or how. |
phantom is still failing on port 4444 however |
Chrome on linux seems to have stopped working. Can't find any relevant bug reports, and reverting to older chromedrivers and selenium-standalone versions didn't seem to make any difference, so perhaps it's something else in the environment? It's possible that it's a timing issue and selenium isn't ready? Switched to use firefox for travis PR tests rather than chrome, for now. 62511c0 |
There's still an intermittent problem where the R system tests timeout at 60 sec. The initial R system test sometimes takes upwards of 30 seconds (witnessed on my system) |
Travis is passing at the moment. Both chrome/linux and R issues seem to have gone away. |
No description provided.
The text was updated successfully, but these errors were encountered: