Upgrade rq 1.3.0 -> 1.5.2

[petlack]

Fixes #362

[waldner]

While this fixes the original issue, I now get:

rq-dashboard    | [2020-10-03 14:47:35,257] ERROR in app: Exception on /0/data/jobs/default/failed/11/1.json [GET]
rq-dashboard    | Traceback (most recent call last):
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/app.py", line 2447, in wsgi_app
rq-dashboard    |     response = self.full_dispatch_request()
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/app.py", line 1952, in full_dispatch_request
rq-dashboard    |     rv = self.handle_user_exception(e)
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/app.py", line 1821, in handle_user_exception
rq-dashboard    |     reraise(exc_type, exc_value, tb)
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/_compat.py", line 39, in reraise
rq-dashboard    |     raise value
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/app.py", line 1950, in full_dispatch_request
rq-dashboard    |     rv = self.dispatch_request()
rq-dashboard    |   File "/usr/local/lib/python3.8/site-packages/flask/app.py", line 1936, in dispatch_request
rq-dashboard    |     return self.view_functions[rule.endpoint](**req.view_args)
rq-dashboard    |   File "/rq_dashboard/web.py", line 101, in _wrapped
rq-dashboard    |     result_dict = f(*args, **kwargs)
rq-dashboard    |   File "/rq_dashboard/web.py", line 427, in list_jobs
rq-dashboard    |     total_items, jobs = get_queue_registry_jobs_count(
rq-dashboard    |   File "/rq_dashboard/web.py", line 234, in get_queue_registry_jobs_count
rq-dashboard    |     jobs = [serialize_job(job) for job in current_queue_jobs]
rq-dashboard    |   File "/rq_dashboard/web.py", line 234, in <listcomp>
rq-dashboard    |     jobs = [serialize_job(job) for job in current_queue_jobs]
rq-dashboard    |   File "/rq_dashboard/web.py", line 169, in serialize_job
rq-dashboard    |     id=job.id,
rq-dashboard    | AttributeError: 'NoneType' object has no attribute 'id'

when I try to look at failed jobs.

[jmandt]

@mgax @nvie would one of you guys be able to have a look at the pull request and merge it if this update of rq version is reasonable?

[mgax]

@mgax @nvie would one of you guys be able to have a look at the pull request and merge it if this update of rq version is reasonable?

Hey @jmandt, I haven't used rq-dashboard in years, can't help you there, sorry :)

[Cybernisk]

Hi, @jmandt I've got the same problem as @waldner above with rq==1.5.2 installed inside but with Finished jobs list.

[Cybernisk]

I guess I found the reason for the bug. RQ registries are not cleaned. My example is that I have 29 jobs in FinishedJobRegistry, but if I count all Redis keys rq:job:* it gives me only 8. I'm not sure, but it could be because of the old rq package I use. Seems that it was fixed in this MR. I will update all rq and rq-related packages in my project and will see how it goes.