Skip to content

Conversation

@joechenrh
Copy link
Contributor

@joechenrh joechenrh commented Nov 7, 2025

What problem does this PR solve?

Issue Number: ref #63366

Problem Summary:

If the last sub job in the multi schema change doesn't need reorg (i.e., no index involved), the analyze task will be skipped.

Take the following SQL as example, since c2 has no index on it, the state is directly converted to AnalyzeStateSkipped, even if modifying c1 requires reorg.

create table t2(id int, c1 int, c2 int, index i1(c1));
alter table t2 modify column c1 char(16), modify column c2 char(16);
c1: ReorgStageModifyColumnCompleted -> AnalyzeStateSkipped
(NeedAnalyze == false)
c2: ReorgStageModifyColumnCompleted -> AnalyzeStateSkipped
(NeedAnalyze == true & len(changingIdx) == 0)

What changed and how does it work?

The problems is caused by inaccurate value for SubJob.NeedAnalyze. That is, the last job with NeedAnalyze == true is not the last job that does reorg. To fix this:

  • Replace CtxVars with boolean value, this value is not accurate and is only used in job submission (we can only get the ModifyColumnType during execution)
  • Set the NeedAnalyze before each subjob becoming non-revertible. For MODIFY COLUMN, it's set during ddl execution, while for ADD INDEX, we can just set it during submission.
  • After the whole job become non-revertible, keep only the last value of NeedAnalyze=true
  • Do analyze for the subjob with NeedAnalyze=true
  • AnalyzeStateSkipped is removed, since we need do the check before updating the table meta.

One of the biggest change after this PR is that, the multi-schema-change is marked as non-revertible after all reorg is finish, not analyze is finished.

This PR also do the following things

  1. move a bunch of function, variables related to modify column from ddl to model package.
  2. filter out temporary columns from the display information of analyze. Previously, we may get output like: auto analyze table all indexes, columns a, b, c,_Col$_c_0 with 256 buckets, 100 topn, 1 samplerate

Check List

Tests

  • Unit test
  • Integration test
  • Manual test (add detailed scripts or steps below)
  • No need to test
    • I checked and no code files have been changed.

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

Please refer to Release Notes Language Style Guide to write a quality release note.

None

Signed-off-by: Ruihao Chen <[email protected]>
@ti-chi-bot ti-chi-bot bot added release-note-none Denotes a PR that doesn't merit a release note. needs-cherry-pick-release-8.5 Should cherry pick this PR to release-8.5 branch. labels Nov 7, 2025
@ti-chi-bot
Copy link

ti-chi-bot bot commented Nov 7, 2025

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign benjamin2037 for approval. For more information see the Code Review Process.
Please ensure that each of them provides their approval before proceeding.

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

Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@ti-chi-bot ti-chi-bot bot added the size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files. label Nov 7, 2025
@tiprow
Copy link

tiprow bot commented Nov 7, 2025

Hi @joechenrh. Thanks for your PR.

PRs from untrusted users cannot be marked as trusted with /ok-to-test in this repo meaning untrusted PR authors can never trigger tests themselves. Collaborators can still trigger tests on the PR using /test all.

I understand the commands that are listed here.

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 kubernetes-sigs/prow repository.

@codecov
Copy link

codecov bot commented Nov 7, 2025

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 74.1633%. Comparing base (5215e99) to head (75d5b16).
⚠️ Report is 13 commits behind head on master.

Additional details and impacted files
@@               Coverage Diff                @@
##             master     #64337        +/-   ##
================================================
+ Coverage   72.7418%   74.1633%   +1.4214%     
================================================
  Files          1859       1883        +24     
  Lines        504114     513747      +9633     
================================================
+ Hits         366702     381012     +14310     
+ Misses       115137     108881      -6256     
- Partials      22275      23854      +1579     
Flag Coverage Δ
integration 47.9741% <ø> (?)

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

Components Coverage Δ
dumpling 52.8700% <ø> (ø)
parser ∅ <ø> (∅)
br 62.3013% <ø> (+15.9509%) ⬆️
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

Signed-off-by: Ruihao Chen <[email protected]>
Signed-off-by: Ruihao Chen <[email protected]>
@joechenrh
Copy link
Contributor Author

/ok-to-test

@ti-chi-bot ti-chi-bot bot added the ok-to-test Indicates a PR is ready to be tested. label Nov 7, 2025
Signed-off-by: Ruihao Chen <[email protected]>
@tiprow
Copy link

tiprow bot commented Nov 7, 2025

@joechenrh: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name Commit Details Required Rerun command
fast_test_tiprow 75d5b16 link true /test fast_test_tiprow

Full PR test history. Your PR dashboard.

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 kubernetes-sigs/prow repository. I understand the commands that are listed here.

@ti-chi-bot
Copy link

ti-chi-bot bot commented Nov 7, 2025

@joechenrh: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name Commit Details Required Rerun command
idc-jenkins-ci-tidb/check_dev 75d5b16 link true /test check-dev
pull-unit-test-next-gen 75d5b16 link true /test pull-unit-test-next-gen
idc-jenkins-ci-tidb/unit-test 75d5b16 link true /test unit-test
pull-unit-test-ddlv1 75d5b16 link true /test pull-unit-test-ddlv1

Full PR test history. Your PR dashboard.

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 kubernetes-sigs/prow repository. I understand the commands that are listed here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

needs-cherry-pick-release-8.5 Should cherry pick this PR to release-8.5 branch. ok-to-test Indicates a PR is ready to be tested. release-note-none Denotes a PR that doesn't merit a release note. size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant