-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[runner] Update to Chromedriver 96.0 custom #66
Conversation
WIP still... have to recompile chrome. |
This works around Behat/MinkExtension#376.
Okay, I've worked out what's going on. Selenium 4.X validates paramers, and does not like the I've raised moodlehq/moodle-behat-extension#50 moodlehq/moodle-behat-extension#51 moodlehq/moodle-behat-extension#52 moodlehq/moodle-behat-extension#53 to work around it in our driver. I've also raised Behat/MinkExtension#377 against Behat/MinkExtension and FriendsOfBehat/MinkExtension#10 against the FOB version. Tests above are now passing (except one which is not a symptom of this issue. |
Hi Andrew, nice work! Just sharing that the complete run above, with the new sele-4 image has ended with a couple of the infame zero-size failures. (link. My understanding is that, no matter of that... still this is a good change because will kill the on-the-edge failures. And also enable CI to use sele-4. So I'm going to proceed. Once merged I'll look to the moodle-behat-extensions, tagging them and creating a MDL to update our branches. Finally, about the Mink extension, just commenting about we already using our own fork of the FriendsOfBehat one, because they never applied some PHP8 fixes to it (from memory). So surely we need to apply the same patch to: https://github.com/moodlehq/MinkExtension.git Ciao :-) |
Just had a thought... if we fix the marionette problem in our MinkExtension fork... then surely we don't need to do anything in moodle-behat-extension? |
I've relaunched the Chrome job above, with the new rebuilt once again selenium 96 image. Let's see how it goes this time: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18810/ |
I've just built it against 3.141.59 and kicked off a job for comparison of speed: There's some speed issues reported in SeleniumHQ/docker-selenium#1444 and it looks like S4 is 33% slower! |
Here's a run against Selenium 3.141.59: |
Yay, so it passed! In 3h22m! Little detail, I saw that https://hub.docker.com/r/moodlehq/selenium-standalone-chrome is still saying sele-4 instead of sele-3 (in case we are going to stay with sele-3 finally). For the records, I've launched one more run, this time using the STABLE runner (current) against the very same commit, to confirm that, without the new image, we were getting the failures: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18850/ |
I've created moodlehq/selenium-standalone-chrome#1 to remember about the non-matching README information. |
@stronk7 @andrewnicols I apologize for blocking you ( |
Thanks @Yozhef ! I imagine this implies that we can now stop using our MinkExtension fork and switch back to the FriendsOfBehat one. Ciao :-) |
Switching to the 96.0 image of https://github.com/moodlehq/selenium-standalone-chrome
Note: Selenium no longer produce images for Selenium 3.141.59 so this image is based on Selenium 4.1!
Before: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18727/
After: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18796/
Just affected Scenarios:
Before: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18803
After: https://ci.moodle.org/view/Testing/job/DEV.01%20-%20Developer-requested%20Behat/18804