pd-server v2.1.14

• Adjust the log output level from Error to Warning when reading an invalid configuration item #1577

pd-server v3.0.0

• Support re-creating a cluster from a single node
• Migrate Region metadata from etcd to the go-leveldb storage engine to solve the storage bottleneck in etcd for large-scale clusters
• API
  • Add the remove-tombstone API to clear Tombstone stores
  • Add the ScanRegions API to batch query Region information
  • Add the GetOperator API to query running operators
  • Optimize the performance of the GetStores API
• Configurations
  • Optimize configuration check logic to avoid configuration item errors
  • Add enable-one-way-merge to control the direction of Region merge
  • Add hot-region-schedule-limit to control the scheduling rate for hot Regions
  • Add hot-region-cache-hits-threshold to identify hotspot when hitting multiple thresholds consecutively
  • Add the store-balance-rate configuration item to control the maximum numbers of balance Region operators allowed per minute
• Scheduler Optimizations
  • Add the store limit mechanism for separately controlling the speed of operators for each store
  • Support the waitingOperator queue to optimize the resource race among different schedulers
  • Support scheduling rate limit to actively send scheduling operations to TiKV. This improves the scheduling rate by limiting the number of concurrent scheduling tasks on a single node.
  • Optimize the Region Scatter scheduling to be not restrained by the limit mechanism
  • Add the shuffle-hot-region scheduler to facilitate TiKV stability test in scenarios of poor hotspot scheduling
• Simulator
  • Add simulator for data import scenarios
  • Support setting different heartbeats intervals for the Store
• Others
  • Upgrade etcd to solve the issues of inconsistent log output formats, Leader selection failure in prevote, and lease deadlocking.
  • Develop a unified log format specification with restructured log system to facilitate collection and analysis by tools
  • Add monitoring metrics including scheduling parameters, cluster label information, time consumed by PD to process TSO requests, Store ID and address information, etc.

pd-server v3.0.0-rc.3

• Add the enable-two-way-merge configuration item to allow only one-way merging #1583
• Add scheduling operations for AddLightLearner and AddLightPeer to make Region Scatter scheduling unrestricted by the limit mechanism #1563
• Fix the issue of insufficient reliability because the data might only have one replica replication when the system is started #1581
• Optimize configuration check logic to avoid configuration item errors #1585
• Adjust the definition of the store-balance-rate configuration to the upper limit of the number of balance operators generated per minute #1591
• Fix the issue that the store might have been unable to generate scheduled operations #1590

pd-server v2.1.13

No changes

pd-server v2.1.12

• Fix the issue that etcd leader election is blocked in extreme conditions
  #1576

pd-server v2.1.11

• Fix the issue that hot Region may fail to be scheduled due to balance-region #1551
• Set hotspot related scheduling priorities to high #1551
• Add two configuration items #1551
  • hot-region-schedule-limit to control the maximum number of concurrent hotspot scheduling tasks
  • hot-region-cache-hits-threshold to identify a hotspot Region

pd-server v3.0.0-rc.2

• Enable the Region storage by default to store the Region metadata #1524
• Fix the issue that hot Region scheduling is preempted by another scheduler #1522
• Fix the issue that the priority for the leader does not take effect #1533
• Add the gRPC interface for ScanRegions #1535
• Push operators actively #1536
• Add the store limit mechanism for separately controlling the speed of operators for each store #1474
• Fix the issue of inconsistent Config status #1476

pd-server v2.1.10

• Fix the issue that the priority of the leader does not take effect #1533

pd-server v2.1.9

• Support the GetOperator service #1514
• Improve leader election related logs #1529

pd-server v3.0.0-rc.1

• Upgrade ETCD #1452
  • Unify the log format of etcd and PD server
  • Fix the issue of failing to elect Leader by PreVote
  • Support fast dropping the “propose” and “read” requests that are to fail to avoid blocking the subsequent requests
  • Fix the deadlock issue of Lease

• Fix the issue that a hot store makes incorrect statistics of keys #1487
• Support forcibly rebuilding a PD cluster from a single PD node
• #1485
• Fix the issue that regionScatterer might generate an invalid OperatorStep
• #1482
• Fix the too short timeout issue of the MergeRegion operator #1495
• Support giving high priority to hot region scheduling #1492
• Add the metrics for recording the time of handling TSO requests on the PD server side #1502
• Add the corresponding Store ID and Address to the metrics related to the store #1507
• Support the GetOperator service #1477
• Fix the issue that the error cannot be sent in the Heartbeat stream because the store cannot be found #1522