zimulala
Follow
Popular repositories
-
reading-go Public
Forked from talkgo/night
Go 夜读 > 每周通过 zoom 在线直播的方式分享 Go 相关的技术话题，每天大家在 微信/Slack 上及时沟通交流编程技术话题。
-
builddatabase Public
Forked from ngaut/builddatabase
Build a distributed SQL database from the ground up
-
-
-
573 contributions in the last year
Contribution activity
February 2023
Created 12 commits in 2 repositories
Created a pull request in pingcap/tidb that received 24 comments
ddl: fix owner get panic info with dist-reorg
What problem does this PR solve? Issue Number: close #41208
Problem Summary:
What is changed and how it works?
Add references to record thereorgCtx …
+77
−5
•
24
comments
Opened 15 other pull requests in 2 repositories
pingcap/tidb
3
closed
2
open
8
merged
- distribute_framework: implement the simplest logical flow for dispatcher
- *: revert-2 for test
- ddl: for test
- *: revert-1 for test
- ddl: move some dist-reorg code to dist_* files and add the generate prefix key function
- *: revert for test
- ddl: fix double/float data not being truncated for column modification operations when the number of decimal places is reduced.
-
ddl: make
reorgCtxsafer to use - ddl: fix cleanup subtasks after cancel add index with dist-reorg
- docs: add a proposal for distributed DDL reorg
-
util/gpool/spmc: add a sleep to a simple
for - ddl, model: support for dist-reorg on partitioned tables
- ddl: remove useless error log
Reviewed 29 pull requests in 3 repositories
pingcap/tidb
25 pull requests
- ddl: Increase ddl unit test speed
- ddl: remove unused field and clean comments
- ddl: refine the error handling for runReorgJob
- ddl: fix unexpected duplicated error when adding unique index (#41687)
- ddl: move some dist-reorg code to dist_* files and add the generate prefix key function
- ddl: use a param to decide whether exec in distributed mode (#41549)
- ddl: use a param to decide whether exec in distributed mode
- partition: check global index when execute create table
- ddl: correct the remain ranges check for adding index (#41460)
- ddl: correct the remain ranges check for adding index (#41460)
- ddl: correct the remain ranges check for adding index
-
ddl: make
reorgCtxsafer to use - *: improve makefile
- ddl: fix cleanup subtasks after cancel add index with dist-reorg
- ddl: improve backfill more general (#41093)
- ddl: Add telementry for distributed reorg tasks. (#41201)
- ddl: improve backfill more general
- bazel: set --test_keep_going false
- *: remove limit
- ddl: Add telementry for distributed reorg tasks.
- ddl, model: support for dist-reorg on partitioned tables
-
store/copr: add a param "limit" to region cache's
SplitRegionRanges(#40411) -
session: use
mDDLTableVersionkey to control backfill tables - resourcemanager: fix TaskController.Stop() can't make producer exit in spmcpool
- ddl, parser: implement the normal flow of distributed processing in DDL reorg phase for adding index
- Some pull request reviews not shown.
pingcap/docs-cn
2 pull requests
pingcap/docs
1 pull request
Created an issue in pingcap/tidb that received 1 comment
Opened 14 other issues in 1 repository
pingcap/tidb
5
open
9
closed
- Implement the simplest logical flow for dispatcher
-
The
likeinRemoveBackfillJobin job_table.go is incorrect - Move some dist-reorg code to dist_* files and add the generate prefix key function
- Unified Distributed Parallel Execution Framework
- Unstable test: TestUsingReorgCtx
- It mayn't cleanup subtasks after cancel add index with dist-reorg
- Data race in TTL
- Return a panic error when other TiDBs exit earlier than the owner
- Support for dist-reorg on partitioned tables
-
Delete unnecessary error logs with
TiDBEnableDistributeReorg -
Call
TaskController.Stop()can't make producer exit in spmcpool - CI failed due to a potential deadlock in statistics/handle
- Change the table structure to be more general
-
Adjust the number of workers corresponding to
backfillWorkerPoolby adjustingtidb_ddl_reorg_worker_cnt
1
contribution
in private repositories
Feb 10