TiDB Operator 1.0 Beta.2 Release Notes

Release date: May 10, 2019

TiDB Operator version: 1.0.0-beta.2

v1.0.0-beta.2 What’s New

Stability has been greatly enhanced

  • Refactored e2e test
  • Added stability test, 7x24 running

Greatly improved ease of use

  • One-command deployment for AWS, Aliyun
  • Minikube deployment for testing
  • Tkctl cli tool
  • Refactor backup chart for ease use
  • Refine initializer job
  • Grafana monitor dashboard improved, support multi-version
  • Improved user guide
  • Contributing documentation

Bug fixes

  • Fix PD start script, add join file when startup
  • Fix TiKV failover take too long
  • Fix PD ha when replcias is less than 3
  • Fix a tidb-scheduler acquireLock bug and emit event when scheduled failed
  • Fix scheduler ha bug with defer deleting pods
  • Fix a bug when using shareinformer without deepcopy

Other improvements

  • Remove pushgateway from TiKV pod
  • Add GitHub templates for issue reporting and PR
  • Automatically set the scheduler K8s version
  • Switch to go module
  • Support slow log of TiDB

Detailed Bug Fixes and Changes

  • Don’t initialize when there is no tidb.password (#282)
  • Fix join script (#285)
  • Document tool setup and e2e test detail in CONTRIBUTING.md (#288)
  • Update setup.md (#281)
  • Support slow log tailing sidcar for TiDB instance (#290)
  • Flexible tidb initializer job with secret set outside of helm (#286)
  • Ensure SLOW_LOG_FILE env variable is always set (#298)
  • Fix setup document description (#300)
  • Refactor backup (#301)
  • Abandon vendor and refresh go.sum (#311)
  • Set the SLOW_LOG_FILE in the startup script (#307)
  • Automatically set the scheduler K8s version (#313)
  • TiDB stability test main function (#306)
  • Add fault-trigger server (#312)
  • Add ad-hoc backup and restore function (#316)
  • Add scale & upgrade case functions (#309)
  • Add slack (#318)
  • Log dump when test failed (#317)
  • Add fault-trigger client (#326)
  • Monitor checker (#320)
  • Add blockWriter case for inserting data (#321)
  • Add scheduled-backup test case (#322)
  • Port ddl test as a workload (#328)
  • Use fault-trigger at e2e tests and add some log (#330)
  • Add binlog deploy and check process (#329)
  • Fix e2e can not make (#331)
  • Multi TiDB cluster testing (#334)
  • Fix backup test bugs (#335)
  • Delete blockWrite.go and use blockwrite.go instead (#333)
  • Remove vendor (#344)
  • Add more checks for scale & upgrade (#327)
  • Support more fault injection (#345)
  • Rewrite e2e (#346)
  • Add failover test (#349)
  • Fix HA when the number of replicas are less than 3 (#351)
  • Add fault-trigger service file (#353)
  • Fix dind doc (#352)
  • Add additionalPrintColumns for TidbCluster CRD (#361)
  • Refactor stability main function (#363)
  • Enable admin privilege for prom (#360)
  • Update README.md with new info (#365)
  • Build CLI (#357)
  • Add extraLabels variable in tidb-cluster chart (#373)
  • Fix TiKV failover (#368)
  • Separate and ensure setup before e2e-build (#375)
  • Fix codegen.sh and lock related dependencies (#371)
  • Add sst-file-corruption case (#382)
  • Use release name as default clusterName (#354)
  • Add util class to support adding annotations to Grafana (#378)
  • Use Grafana provisioning to replace dashboard installer (#388)
  • Ensure test env is ready before cases running (#386)
  • Remove monitor config job check (#390)
  • Update local-pv documentation (#383)
  • Update Jenkins links in README.md (#395)
  • Fix e2e workflow in CONTRIBUTING.md (#392)
  • Support running stability test out of cluster (#397)
  • Update TiDB secret docs and charts (#398)
  • Enable blockWriter write pressure in stability test (#399)
  • Support debug and ctop commands in CLI (#387)
  • Marketplace update (#380)
  • Update editable value from true to false (#394)
  • Add fault inject for kube proxy (#384)
  • Use ioutil.TempDir() create charts and operator repo’s directories (#405)
  • Improve workflow in docs/google-kubernetes-tutorial.md (#400)
  • Support plugin start argument for TiDB instance (#412)
  • Replace govet with official vet tool (#416)
  • Allocate 24 PVs by default (after 2 clusters are scaled to (#407)
  • Refine stability (#422)
  • Record event as grafana annotation in stability test (#414)
  • Add GitHub templates for issue reporting and PR (#420)
  • Add TiDBUpgrading func (#423)
  • Fix operator chart issue (#419)
  • Fix stability issues (#433)
  • Change cert generate method and add pd and kv prestop webhook (#406)
  • A tidb-scheduler bug fix and emit event when scheduled failed (#427)
  • Shell completion for tkctl (#431)
  • Delete an duplicate import (#434)
  • Add etcd and kube-apiserver faults (#367)
  • Fix TiDB Slack link (#444)
  • Fix scheduler ha bug (#443)
  • Add terraform script to auto deploy TiDB cluster on AWS (#401)
  • Add instructions to access Grafana in GKE tutorial (#448)
  • Fix label selector (#437)
  • No need to set ClusterIP when syncing headless service (#432)
  • Document how to deploy TiDB cluster with tidb-operator in minikube (#451)
  • Add slack notify (#439)
  • Fix local dind env (#440)
  • Add terraform scripts to support alibaba cloud ACK deployment (#436)
  • Fix backup data compare logic (#454)
  • Async emit annotations (#438)
  • Use TiDB v2.1.8 by default & remove pushgateway (#435)
  • Fix a bug that uses shareinformer without copy (#462)
  • Add version command for tkctl (#456)
  • Add tkctl user manual (#452)
  • Fix binlog problem on large scale (#460)
  • Copy kubernetes.io/hostname label to PVs (#464)
  • AWS EKS tutorial change to new terraform script (#463)
  • Update documentation of minikube installation (#471)
  • Update documentation of DinD installation (#458)
  • Add instructions to access Grafana (#476)
  • Support-multi-version-dashboard (#473)
  • Update Aliyun deploy docs after testing (#474)
  • GKE local SSD size warning (#467)
  • Update roadmap (#376)