fit: optimize the selection of witness rule candidates

[ethercflow]

ref #5568

Signed-off-by: Wenbo Zhang [email protected]

What problem does this PR solve?

Issue Number: ref #5568

What is changed and how does it work?

1. If the IsWitness of the Rule is true, the leader will be removed from the candidate list, so that when the best rule is selected, it is impossible for the leader to exist in the witness rule;
2. When sorting the candidate peers, if the two peers participating in the comparison belong to the same storestate, then 50% of them will be randomly sorted to prevent the sorted peer list from being fixed, so that when the best rule is selected, the fixed peer will always be returned, causing some store balance, a store is 0;

before:

After:

Update:
If the rule remains unchanged, each time the rule_checker checks the region, it will directly fetch the previously calculated best rule from the cache without being randomly affected. But If the rule is manually modified, the cache will become invalid, and the rule_checker will recalculate each region, which may cause unnecessary scheduling for existing peers, such as adding TiFlash. So remove the random strategy.

Check List

Tests

• None

Code changes

• None

Side effects

• None

Related changes

• None

Release note

None

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has been approved by:

• bufferflies
• rleungx

To complete the pull request process, please ask the reviewers in the list to review by filling /cc @reviewer in the comment.
After your PR has acquired the required number of LGTMs, you can assign this pull request to the committer in the list by filling /assign @committer in the comment to help you merge this pull request.

The full list of commands accepted by this bot can be found here.

Reviewer can indicate their review by submitting an approval review.
Reviewer can cancel approval by submitting a request changes review.

[wuhuizuo]

/test build

[wuhuizuo]

/test build

[ethercflow]

Random might cause some test cases to fail, I'll think again (:

[codecov]

Codecov Report

Base: 75.46% // Head: 75.49% // Increases project coverage by +0.02% 🎉

Coverage data is based on head (73d253a) compared to base (03081c6).
Patch coverage: 100.00% of modified lines in pull request are covered.

❗ Current head 73d253a differs from pull request most recent head cb30f33. Consider uploading reports for the commit cb30f33 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #5904      +/-   ##
==========================================
+ Coverage   75.46%   75.49%   +0.02%     
==========================================
  Files         345      345              
  Lines       35132    35126       -6     
==========================================
+ Hits        26514    26517       +3     
+ Misses       6322     6319       -3     
+ Partials     2296     2290       -6     

Flag	Coverage Δ
unittests	75.49% <100.00%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
server/schedule/placement/fit.go	95.83% <100.00%> (ø)
pkg/errs/errs.go	75.00% <0.00%> (-25.00%)	⬇️
pkg/utils/tempurl/tempurl.go	45.00% <0.00%> (-15.00%)	⬇️
pkg/id/id.go	83.05% <0.00%> (-3.39%)	⬇️
pkg/member/member.go	64.21% <0.00%> (-1.58%)	⬇️
server/region_syncer/client.go	86.56% <0.00%> (-1.50%)	⬇️
tools/pd-ctl/pdctl/command/operator.go	66.66% <0.00%> (-1.15%)	⬇️
server/region_syncer/server.go	81.86% <0.00%> (-1.10%)	⬇️
server/cluster/coordinator.go	71.59% <0.00%> (-0.76%)	⬇️
pkg/tso/tso.go	70.22% <0.00%> (-0.57%)	⬇️
... and 18 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[bufferflies]

/merge

[ti-chi-bot]

@bufferflies: It seems you want to merge this PR, I will help you trigger all the tests:

/run-all-tests

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository.

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 73d253a

[ti-chi-bot]

@ethercflow: Your PR was out of date, I have automatically updated it for you.

If the CI test fails, you just re-trigger the test that failed and the bot will merge the PR for you after the CI passes.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository.