rl/e2e configure orchestrator

[red-0ne]

What does this pull request do? Explain your changes. (required)
E2E test orchestrator configuration

Specific updates (required)

• Start a geth node
• Start go-livepeer node in orchestrator mode
• Register orchestrator
• Wait for the next round
• Reconfigure orchestrator
• Check updated values

How did you test each of these updates (required)
Tests with ./test.sh and ./test_e2e.sh pass

Does this pull request close any open issues?
Fixes #2421

Checklist:

• Read the contribution guide
• make runs successfully
• All tests in ./test.sh pass
• README and other documentation updated
• Pending changelog updated

[codecov]

Codecov Report

Merging #2443 (f1c239a) into master (2e9387e) will decrease coverage by 0.03218%.
The diff coverage is 50.00000%.

@@                 Coverage Diff                 @@
##              master       #2443         +/-   ##
===================================================
- Coverage   55.05601%   55.02383%   -0.03219%     
===================================================
  Files             94          94                 
  Lines          19729       19726          -3     
===================================================
- Hits           10862       10854          -8     
- Misses          8265        8270          +5     
  Partials         602         602                 

Impacted Files	Coverage Δ
cmd/livepeer/starter/starter.go	2.33236% <0.00000%> (-0.00454%)	⬇️
server/webserver.go	95.69892% <100.00000%> (-0.21945%)	⬇️
discovery/db_discovery.go	67.01389% <0.00000%> (-1.04167%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2e9387e...f1c239a. Read the comment docs.

[leszko]

Thanks for the PR Redouane. I added a few comments. Other than that:

1. Could you update the Checklist in the PR description (either check the box or cross out the whole point if it's not related)
2. There is something weird about why your test takes so much time to execute and why you had to split it into 2 tests.

I tried to do all the required steps manually using devtool and livepeer_cli. And doing them manually took only 1.5 min. So, I guess your test should take no more than 1.5 min. I recored a movie of the steps. Could you please check and try it yourself and then try to see what's wrong with the test you wrote?

2022-06-15.12-44-32.mp4

[red-0ne]

Could you please check and try it yourself and then try to see what's wrong with the test you wrote?

I think I know what's happening. It's actually when changing blockRewardCut and feeShare that are on-chain operations that require waiting for rounds. I guess if you try to trigger option 13. Set orchestrator config after being registered and activated, this will cause the error I faced. Will double check it and update here.

[leszko]

Thanks @red-0ne for all the work. Added some more comments. They are mainly about:

1. Removing the time-consuming withdraw stake part
2. Some restructure of the tests for the sake of readability

Other than that, the tests are fine. I'm happy we have them :)

[leszko]

Added one comment. Other than that LGTM 👍

Thanks for the good work @red-0ne