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
The issue with discovery retirement is located here.
[Quick solution] Remove the Discovery step from the pipeline.
That's because the config is updated with services' APIs here, but there is no Discovery service -> we are getting the key error when trying to get it from the config -> pipeline fails. In other words - tubular can't (and never could) handle Discovery retirement.
The same story for the retirement scripts migrated to the edx-platform.
[Better one] Update the retirement scripts in the edx-platform (tubular is dead) to support discovery retirement.
RETIRE_MAILINGS pipeline step should be removed from the plugin as well because it was deprecated and removed as a separate API endpoint from the LMS starting from the Ironwood release and now it is a part of the LMS_MISC step.
Retirement steps order is different from retirement state indexes in the DB after migration. To avoid errors we need to place Enrollments retirement before the Forum retirement.
What I did
I have a working version that has already been tested on our instances.
Notes:
I also removed the Ecommerce-related steps from the RETIREMENT_STATES list because it is unneeded after the recent PR has been merged.
I applied the "Quick" solution for the Discovery issue which is not the best choice
What I expected to happen
It would be great to apply fixes for the listed issues.
My environment
Please add some information about the environment that you’re working in. At a minimum, include these items:
Tutor version: v17.0.1
tutor-contrib-retirement version: my fork is based on v3.3.1
Output of the retirement command/CronJob:
Discovery error:
Retiring tmp1
b'Learner Retirement: Starting learner retirement for tmp1 using config file ./pipeline_config/config.yml'
b'Learner Retirement: Starting state RETIRING_DISCOVERY'
b"Learner Retirement: Error in retirement state RETIRING_DISCOVERY: 'DISCOVERY'"
b'Learner Retirement: Error encountered in state "RETIRING_DISCOVERY"\n\'DISCOVERY\''
b'Learner Retirement: Traceback (most recent call last):\n File "scripts/retire_one_learner.py", line 193, in retire_learner\n response = getattr(config[service], method)(learner)\nKeyError: \'DISCOVERY\'\n'
For wrong pipeline steps order:
Moved from FORUMS_COMPLETE (step 81) to ERRORED (step 131):
400 Client Error: Bad Request for url: https://lms-philantrophy-dev.raccoongang.net/api/user/v1/accounts/update_retirement_status/
The text was updated successfully, but these errors were encountered:
@dyudyunov Sorry for the slow response here; annual leave got in the way. I've taken the changes from your topic branch, broken them into 4 commits instead of 2, updated some of the commit messages, and added updates to the documentation. See PR #40.
@dyudyunov Sorry for the slow response here; annual leave got in the way. I've taken the changes from your topic branch, broken them into 4 commits instead of 2, updated some of the commit messages, and added updates to the documentation. See PR #40.
Summary
[Quick solution] Remove the Discovery step from the pipeline.
That's because the config is updated with services' APIs here, but there is no Discovery service -> we are getting the key error when trying to get it from the config -> pipeline fails. In other words - tubular can't (and never could) handle Discovery retirement.
The same story for the retirement scripts migrated to the edx-platform.
[Better one] Update the retirement scripts in the edx-platform (tubular is dead) to support discovery retirement.
What I did
I have a working version that has already been tested on our instances.
Notes:
What I expected to happen
It would be great to apply fixes for the listed issues.
My environment
Please add some information about the environment that you’re working in. At a minimum, include these items:
tutor-contrib-retirement
version: my fork is based on v3.3.1Discovery error:
For wrong pipeline steps order:
The text was updated successfully, but these errors were encountered: