owner: fix bug of owner manager hang

[tiancaiamao]

What problem does this PR solve?

Issue Number: close #58620

Problem Summary:

What changed and how does it work?

The owner compaign loop use wrong ctx for refressSession, so owner.Close() may hang sometimes.

Before the fix, the behavior is:

ctx := context.Background()
m := NewOwnerManager(ctx, ...)
m.Close()  // no, this cannot make m exit

Call Close() may hang, unless we pass a ctx to NewOwnerManager and cancel that ctx:

ctx, cancel := context.WithCancel(context.Background())
m := NewOwnerManager(ctx, ...)
m.Close() // no, this cannot make m exit
cancel()  // ok, this can make m exit

This is not what I expected. Close() should not hang!

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

[ti-chi-bot]

[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.

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

Needs approval from an approver in each of these files:

• OWNERS
• pkg/owner/OWNERS

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

[tiprow]

Hi @tiancaiamao. 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.

[tiancaiamao]

This is the core of the fix.
We should pass the campaignContext.

The code is:

campaignContext, cancel := context.WithCancel(m.ctx)
go campaignLoop(campaignContext)
cancel()

If we use m.ctx in the campaign loop, cancel() cannot cancel the ctx!

[tiancaiamao]

Without the fix, the test hang... because the old code use wrong ctx and cancel in owner manager Close() does make campaign loop break.

[ti-chi-bot]

@tiancaiamao: 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
idc-jenkins-ci-tidb/unit-test	83c610c	link	true	/test unit-test

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.

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 13 lines in your changes missing coverage. Please review.

Project coverage is 73.5190%. Comparing base (70393b5) to head (83c610c).
Report is 22 commits behind head on master.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #58622        +/-   ##
================================================
+ Coverage   73.1072%   73.5190%   +0.4117%     
================================================
  Files          1676       1676                
  Lines        463352     463975       +623     
================================================
+ Hits         338744     341110      +2366     
+ Misses       103774     102004      -1770     
- Partials      20834      20861        +27     

Flag	Coverage Δ
integration	42.8645% <0.0000%> (?)

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

Components	Coverage Δ
dumpling	52.6910% <ø> (ø)
parser	∅ <ø> (∅)
br	45.8233% <ø> (+0.0089%)	⬆️

[D3Hunter]

not a bug, we purposed do that, see

tidb/pkg/owner/manager.go

Lines 439 to 444 in 42d4fae

	// Note: we must use manager's context to create session. If we use campaign
	// context and the context is cancelled, the created session cannot be closed
	// as session close depends on the context.
	// One drawback is that when you want to break the campaign loop, and the campaign
	// loop is refreshing the session, it might wait for a long time to return, it
	// should be fine as long as network is ok, and acceptable to wait when not.

it's ok to wait, except on shutdown where the context should be cancelled

[D3Hunter]

/hold