make @defnull's proposition into a PR #476

einhirn · 2021-03-04T10:19:42Z

I applied @defnull's code proposition from #108 to current master. Maybe this time the PR will be accepted.

einhirn · 2021-03-04T11:25:26Z

Sorry - I don't understand rubocop's error message. Can someone with ruby knowledge help me fix this?

Nevermind I've got it now. Missed a line when copy&pasting - that's why lint couldn't make heads&tails of the code.

defnull · 2021-03-04T11:34:36Z

We had this in production, but removed it again because it did not behave well in the situation described by follow-up comments. Also, my code sucks, I'm not a ruby developer.

einhirn · 2021-03-04T11:38:56Z

We had this in production, but removed it again because it did not behave well in the situation described by follow-up comments. Also, my code sucks, I'm not a ruby developer.

Combined with the newly added "assume configurable base load for the first few minutes"-feature it could work a bit better.

BTW: I don't know how to better attribute you, other than mentioning...

einhirn · 2021-03-04T11:56:17Z

I guess this should also use
config.x.variable_name
instead of fetching Environment with every poll?

[x] done.

defnull · 2021-03-04T12:08:23Z

BTW: I don't know how to better attribute you, other than mentioning...

No worries, I don't care much about attribution here. I also signed a CLA so this PR being based on my work should not be a problem :)

defnull · 2021-03-04T17:07:28Z

This looks wrong: load += weight_users * total_attendees. The total attendees number grows with each meeting, so you're counting users multiple times.

einhirn · 2021-03-04T21:05:37Z

This looks wrong: load += weight_users * total_attendees. The total attendees number grows with each meeting, so you're counting users multiple times.

Thanks for pointing that out - I'll adress that issue.

[x] done. (having a hard time getting the indentation right.)