-
Notifications
You must be signed in to change notification settings - Fork 6.4k
Checkpoints
Checkpoint is a feature in RocksDB which provides the ability to take a snapshot of a running RocksDB database in a separate directory. Checkpoints can be used as a point in time snapshot, which can be opened Read-only to query rows as of the point in time or as a Writeable snapshot by opening it Read-Write. Checkpoints can be used for both full and incremental backups, as long as the backups are kept on the same device as the DB. (See also How to backup RocksDB)
The Checkpoint feature enables RocksDB to create a consistent snapshot of a given RocksDB database in the specified directory. If the snapshot is on the same filesystem as the original database, the SST files will be hard-linked, otherwise SST files will be copied. The manifest and CURRENT files will be copied. In addition, if there are multiple column families, log files will be copied for the period covering the start and end of the checkpoint, in order to provide a consistent snapshot across column families.
A Checkpoint object needs to be created for a database before checkpoints are created. The API is as follows:
Status Create(DB* db, Checkpoint** checkpoint_ptr);
Given a checkpoint object and a directory, the CreateCheckpoint function creates a consistent snapshot of the database in the given directory.
Status CreateCheckpoint(const std::string& checkpoint_dir);
The directory should not already exist and will be created by this API. The directory will be an absolute path. The checkpoint can be used as a read-only copy of the DB or can be opened as a standalone DB. When opened read/write, the SST files continue to be hard links and these links are removed when the files are obsoleted. When the user is done with the snapshot, the user can delete the directory to remove the snapshot.
Checkpoints are used for online backup in MyRocks, which is MySQL using RocksDB as the storage engine. (MySQL on RocksDB)
Contents
- RocksDB Wiki
- Overview
- RocksDB FAQ
- Terminology
- Requirements
- Contributors' Guide
- Release Methodology
- RocksDB Users and Use Cases
- RocksDB Public Communication and Information Channels
-
Basic Operations
- Iterator
- Prefix seek
- SeekForPrev
- Tailing Iterator
- Compaction Filter
- Multi Column Family Iterator
- Read-Modify-Write (Merge) Operator
- Column Families
- Creating and Ingesting SST files
- Single Delete
- Low Priority Write
- Time to Live (TTL) Support
- Transactions
- Snapshot
- DeleteRange
- Atomic flush
- Read-only and Secondary instances
- Approximate Size
- User-defined Timestamp
- Wide Columns
- BlobDB
- Online Verification
- Options
- MemTable
- Journal
- Cache
- Write Buffer Manager
- Compaction
- SST File Formats
- IO
- Compression
- Full File Checksum and Checksum Handoff
- Background Error Handling
- Huge Page TLB Support
- Tiered Storage (Experimental)
- Logging and Monitoring
- Known Issues
- Troubleshooting Guide
- Tests
- Tools / Utilities
-
Implementation Details
- Delete Stale Files
- Partitioned Index/Filters
- WritePrepared-Transactions
- WriteUnprepared-Transactions
- How we keep track of live SST files
- How we index SST
- Merge Operator Implementation
- RocksDB Repairer
- Write Batch With Index
- Two Phase Commit
- Iterator's Implementation
- Simulation Cache
- [To Be Deprecated] Persistent Read Cache
- DeleteRange Implementation
- unordered_write
- Extending RocksDB
- RocksJava
- Lua
- Performance
- Projects Being Developed
- Misc