server: clean up code for better understanding #27877
Conversation
|
[REVIEW NOTIFICATION] This pull request has been approved by:
To complete the pull request process, please ask the reviewers in the list to review by filling The full list of commands accepted by this bot can be found here. Reviewer can indicate their review by submitting an approval review. |
ti-chi-bot
added
release-note-none
Signed-off-by: dragonly <[email protected]>
dragonly
force-pushed
the
dragonly/clean-up-code
branch
from
fc9968b to
3c16f5e
Compare
server/conn.go
Outdated
| _, allowTiFlashFallback := cc.ctx.GetSessionVars().AllowFallbackToTiKV[kv.TiFlash] | ||
| if allowTiFlashFallback && errors.ErrorEqual(err, storeerr.ErrTiFlashServerTimeout) && retryable { | ||
| if allowTiFlashFallback { |
There was a problem hiding this comment.
The code logic changes after the modification. If allowTiFlashFallback is false, we still need to jump out of the loop.
There was a problem hiding this comment.
Thanks, fixed in the next commit. The main idea is using guard pattern to prevent deep nest of if branches.
server/conn.go
Outdated
| }() | ||
| delete(cc.ctx.GetSessionVars().IsolationReadEngines, kv.TiFlash) | ||
| _, err = cc.handleStmt(ctx, stmt, warns, i == len(stmts)-1) | ||
| cc.ctx.GetSessionVars().IsolationReadEngines[kv.TiFlash] = struct{}{} |
There was a problem hiding this comment.
Here we use defer because we should add TiFlash back to IsolationReadEngines if handleStmt panics. Check #22459 (comment). However, It seems the connection will be closed if handleStmt panics so maybe adding TiFlash back to IsolationReadEngines is unnecessary. @qw4990
There was a problem hiding this comment.
@xuyifangreeneyes Oh there's such a hidden logic here...
I think this panic should not be recovered by adding back TiFlash as a valid IsolationReadEngines, because it's not actually recoverable this way, and may cause more undesired side effects.
There was a problem hiding this comment.
@dragonly Can you remove similar defer in another place (*clientConn).handleStmtExecute, thx~
ti-chi-bot
added
status/LGT2
|
/merge |
|
This pull request has been accepted and is ready to merge. Commit hash: f72c457
|
ti-chi-bot
added
the
status/can-merge
|
@dragonly: Your PR was out of date, I have automatically updated it for you.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. |
What problem does this PR solve?
Issue Number: close #xxx
Problem Summary:
What is changed and how it works?
clean up code for better understanding.
Proposal: xxx
What's Changed:
How it Works:
Check List
Tests
Side effects
Documentation
Release note