-
Notifications
You must be signed in to change notification settings - Fork 240
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
Attempt to fix AWS jobstore test #2874
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM.
They try to run on Python 3 now because I dropped the EC2 SSH key dependency for attempting them
a2bd903
to
e5ddd64
Compare
…fuse to progress with it off.
OK, this gets the tests passing again. I think the problem really was being able to do an upload before versioning got enabled properly, and probably something on Amazon's end (maybe just making things faster?) made it so we started having that happen with appreciable frequency. If we don't see the same problem again, we can eventually remove some of the logging and polling I've added here, and keep just the initial wait for versioning to take effect. |
* Add more logging to try and diagnose cases where version and content disappear * Allow AWS job store tests to run when you don't have EC2 * Add comment * Try and make AWS job store tests run through on Python 3 They try to run on Python 3 now because I dropped the EC2 SSH key dependency for attempting them * use actually extant isinstance * Decode bytes when we need them as strings * Fix some Python 3 incompatibilities and watch our pipe to make sure the reader runs * Make sure uploads actually happen * Run flaky test first * Add logging and retries to try and get object versions after upload * Get the key the right way * Don't use removed message exception attribute * Poll the versioning setting, give it some time to take effect, and refuse to progress with it off. * Don't len a BytesIO * Do integration tests last again
* Add more logging to try and diagnose cases where version and content disappear * Allow AWS job store tests to run when you don't have EC2 * Add comment * Try and make AWS job store tests run through on Python 3 They try to run on Python 3 now because I dropped the EC2 SSH key dependency for attempting them * use actually extant isinstance * Decode bytes when we need them as strings * Fix some Python 3 incompatibilities and watch our pipe to make sure the reader runs * Make sure uploads actually happen * Run flaky test first * Add logging and retries to try and get object versions after upload * Get the key the right way * Don't use removed message exception attribute * Poll the versioning setting, give it some time to take effect, and refuse to progress with it off. * Don't len a BytesIO * Do integration tests last again
This is my attempt to fix #2873. Right now I can't replicate the failure locally, so I've added some more specific errors/logging to see if I can get more information on failures on Gitlab.