variable: make backoff weight could be 0

[jackysp]

Signed-off-by: Shuaipeng Yu [email protected]

What problem does this PR solve?

Issue Number: close #32592

Problem Summary:
TiDB always uses backoff to resolve some internal errors, e.g., node failures, but in some cases a fast failure may be a better choice because it will give the user a clear ack quickly, rather than looking like it hung.

What is changed and how it works?

Allow tidb_backoff_weight to be set to 0 to turn off backoff mechanism.

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)

1. run some sysbench benchmark on a cluster and set global tidb_backoff_weight = 0
2. the cluster run as normal with no node failure.

• No code

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

None

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has been approved by:

• bb7133
• morgo

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.

[jackysp]

/run-unit-test

[djshow832]

Please add tests to verify that the retrying mechanism still works fine when the weight is set to 0.

[sre-bot]

Code Coverage Details: https://codecov.io/github/pingcap/tidb/commit/5969635f7a8a205a3be119db8cf70be213efb499

[jackysp]

Please add tests to verify that the retrying mechanism still works fine when the weight is set to 0.

Setting it to 0 actually disables the retrying mechanism, so it should not work. I actually tested the benchmark test manually with Sysbench for data import and oltp workload, and the cluster behaved normally with no node failures. @djshow832

[jackysp]

PTAL @djshow832 @bb7133

[jackysp]

PTAL @djshow832 @bb7133

[bb7133]

LGTM

[bb7133]

/merge

[ti-chi-bot]

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

Commit hash: df7a73f

[ti-chi-bot]

@jackysp: Your PR was out of date, I have automatically updated it for you.

At the same time I will also trigger all tests for you:

/run-all-tests

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.