Test Pilot Stage and Dev installed add-ons are older than Production add-on based on the version number #2865
Comments
I can reproduce this and will look into it today |
Possibly a lead: b7d0e75#diff-01dd223246b3d82a911250aceaac85b5L95 We stopped uploading Also, tried to switch dev & stage to use local unsigned build of addon (which we're no longer uploading): e60e84c |
- For dev & stage only: - Restore the *.xpi and *.rdf upload removed in mozilla#2651 - Restore creation of /static/addon/latest redirect with add-on hash - Configure frontend app to install add-on from /static/addon/latest Issue mozilla#2865
- For dev & stage only: - Restore the *.xpi and *.rdf upload removed in mozilla#2651 - Restore creation of /static/addon/latest redirect with add-on hash - Configure frontend app to install add-on from /static/addon/latest Issue mozilla#2865
After the latest changes that landed, I have retest this issue using all the 5 prefs and it seems that the behavior is changed. But I am not sure if it's the correct one: Test Pilot dev server: Test Pilot stage server: I'm not sure if it's ok to wait for an auto-update when you install TxP from scratch. Test Pilot production server: I'm not sure if this will affect the production deploy, but like I mentioned in the issue, usually the production version is lower then stage and dev versions. Now on stage and production we have the same version (2.0.2). |
👍
There was a change to the staging deployment process merged into ops' configuration scripts. But, I'm not sure whether that itself deploys automatically, or if we have to wait for an ops deploy cycle. (/cc @relud ?)
For what it's worth, updates are always served up from production. This is not affected by the location from which you installed the add-on. The production update URL is baked into the add-on and can't really vary between environments.
This is as expected, because we haven't pushed to production yet. |
I can confirm that stage still has the old version of the add-on though 😞
|
deploys made after a merge pick up changes to ops' deploy scripts, so the deploy yesterday passed the |
We just did some troubleshooting and I think we've got this resolved on stage!
|
[Notes]:
[Affected versions]:
[Affected Platforms]:
[Prerequisites]:
[Steps to reproduce]:
[Expected result]:
[Actual results]:
[Additional Notes]:
The text was updated successfully, but these errors were encountered: