[Merged by Bors] - reduce hare commitee to 50

[dshulyak]

this will reduce total gossipsub traffix x8.
comitee upgrade is scheduled to be used from layer 105_720 (July 15, 2024, 10:00:00 AM UTC).

[codecov]

Codecov Report

Attention: Patch coverage is 76.92308% with 3 lines in your changes missing coverage. Please review.

Project coverage is 81.5%. Comparing base (2e10983) to head (4e7d38e).

Files	Patch %	Lines
hare3/hare.go	62.5%	2 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff            @@
##           develop   #5793     +/-   ##
=========================================
- Coverage     81.6%   81.5%   -0.1%     
=========================================
  Files          302     302             
  Lines        32488   32498     +10     
=========================================
+ Hits         26512   26513      +1     
- Misses        4240    4247      +7     
- Partials      1736    1738      +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[fasmat]

can we add this to fastnet config so we can check it in system tests?

[dshulyak]

it looks like i missed 2 places in oracle when i created this change.
if you plan to use it please update legacyOracle.go active and validate functions to use CommitteeFor helper

[poszu]

@dshulyak

it looks like i missed 2 places in oracle when i created this change. if you plan to use it please update legacyOracle.go active and validate functions to use CommitteeFor helper

What about the blocks/certifier.go? It currently uses a 200 committee. Can it be larger than Hare's committee size?

[poszu]

bors try

[dshulyak]

What about the blocks/certifier.go? It currently uses a 200 committee. Can it be larger than Hare's committee size?

yes, they are using same algo but otherwise independent.

where 50 came from? doesn't seem to make a difference, so just curious why you decided to change

[poszu]

where 50 came from? doesn't seem to make a difference, so just curious why you decided to change

That's the number we agreed on in the team. There was no specific argument for 50 over 40 🤷.

[poszu]

bors try

[spacemesh-bors]

try

Build succeeded:

• ci-status
• systest-status

[poszu]

bors merge

[spacemesh-bors]

Build failed:

• systest-status

[poszu]

Bors merge

[spacemesh-bors]

Build failed:

• systest-status

[poszu]

bors merge

[spacemesh-bors]

Build failed:

• systest-status

[fasmat]

The failing test indicates that the might be an issue with the change: link

2024-07-06 11:33:30.612 | {"L":"WARN","T":"2024-07-06T09:33:30.612Z","N":"node","M":"certificate committee size is not set, defaulting to hare committee size","size":800,"name":""}

shouldn't this be 50?

There are also errors by the blockGenerator and other components but I'm not sure what's the cause of the problems.

[dshulyak]

there was this failure in 17th layer. perhaps there is some unexpected interaction

2024-07-06 11:41:55.800	
{"L":"WARN","T":"2024-07-06T09:41:55.800Z","N":"node.hare","M":"failed","lid":17,"error":"hare failed to reach consensus in 2 iterations"}


	2024-07-06 11:42:43.566	
{"L":"INFO","T":"2024-07-06T09:42:43.566Z","N":"node.executor","M":"executed empty layer","sessionId":"3274113a","lid":17,"state_hash":"af1349b9f5f9a1a6a0404dea36dcc9499bcb25c9adc112b7cc9a93cae41f3262","duration":"191.612µs","name":"executor"}
	
	
2024-07-06 11:42:43.568	
{"L":"INFO","T":"2024-07-06T09:42:43.567Z","N":"node.executor","M":"executed block","sessionId":"3274113a","lid":18,"block":"48c4afd5b9","state_hash":"d20eb3e63b4be4cdb4b2d37fca0c290ee7b8169b2644cea40e8c97c70be73c1c","duration":"1.161017ms","count":0,"rewards":19,"name":"executor"}
	
	
2024-07-06 11:42:43.575	
{"L":"INFO","T":"2024-07-06T09:42:43.575Z","N":"node.executor","M":"executed block","sessionId":"3274113a","lid":19,"block":"d505834585","state_hash":"6b0728f7687f6349c6823fb1f607debab5ac2d8bd771585293379aa9fe2937f0","duration":"6.99174ms","count":30,"rewards":18,"name":"executor"}
	
	
2024-07-06 11:42:43.580	
{"L":"INFO","T":"2024-07-06T09:42:43.580Z","N":"node.executor","M":"executed block","sessionId":"3274113a","lid":20,"block":"7b6cf0a756","state_hash":"d27ff805a4937aa5f1fb6dd08ca02bbfc0f24881a277ea50a7bd32409dce211e","duration":"4.580587ms","count":12,"rewards":17,"name":"executor"}
	
	
2024-07-06 11:42:50.904	
{"L":"INFO","T":"2024-07-06T09:42:50.904Z","N":"node.executor","M":"executed block","lid":21,"block":"83da1e89e0","state_hash":"fa1a9acd73de7457206fc3b1a6fe6aa0d23e2a0051bea304ec9563b221bf65e8","duration":"2.161088ms","count":0,"rewards":20,"name":"executor"}
	

[dshulyak]

actually it looks accidental. pods were restarted in the middle of the test messages rate

[poszu]

The failing test indicates that the might be an issue with the change: link

2024-07-06 11:33:30.612 | {"L":"WARN","T":"2024-07-06T09:33:30.612Z","N":"node","M":"certificate committee size is not set, defaulting to hare committee size","size":800,"name":""}

shouldn't this be 50?

I was also puzzled by this. That's block certifier committee, not hare. It doesn't need to be equal to the hare committee.

[poszu]

bors try

[spacemesh-bors]

try

Build succeeded:

• ci-status
• systest-status

[poszu]

bors merge

[spacemesh-bors]

Pull request successfully merged into develop.

Build succeeded:

• ci-status
• systest-status