speedb
speedb copied to clipboard
speedb-io
A RocksDB compliant high performance scalable embedded key-value store
With a setting of multiple cfs and WriteBufferManager with allow_stall, the DB can enter a deadlock when the WBM initiates a stall. This happens since only the oldest cf is...
**Describe the bug** db_stress gets stuck **To Reproduce** Steps to reproduce the behavior: 1. rm -rf /tmp/rocksdb_crashtest_* 2. mkdir /tmp/rocksdb_crashtest_blackboxod86ddmf 3. mkdir /tmp/rocksdb_crashtest_expected_vy0z2l_r 4. ./db_stress --acquire_snapshot_one_in=10000 --adaptive_readahead=1 --allow_concurrent_memtable_write=0 --allow_data_in_errors=True --allow_wbm_stalls=1...
After long running of blackbox_crash_test with allow_wbm_stalls = false, at some point there will be an error from the db_stress command such as: stderr: WARNING: prefix_size is non-zero but memtablerep...
Make your vote count : What wrapper do you use today, or what wrapper do you think needs the most work ?
- Report the following per block cache: - Per CF sharing the block cache in the DB, report the following information for the data blocks, index blocks, filter blocks, and...
Bug: https://github.com/facebook/rocksdb/issues/12179 There was one instance where the IsEmpty() check was missing: [rocksdb/db/db_impl/db_impl_write.cc](https://github.com/facebook/rocksdb/blob/21d5a8f54f06e01ca49d4e6bae89bb42ba78dfdd/db/db_impl/db_impl_write.cc#L1706) Line 1706 in [21d5a8f](https://github.com/facebook/rocksdb/commit/21d5a8f54f06e01ca49d4e6bae89bb42ba78dfdd) status = SwitchMemtable(cfd, write_context); Having one check inside SwitchMemTable() will ensure there won't...
**Describe the bug** A clear and concise description of what the bug is. **To Reproduce** Steps to reproduce the behavior: 1. 2. 3. 4. **Expected behavior** A clear and concise...
RocksDB/Speedb has several knobs that can be controlled via the different Options classes and Options files. Different programs and use cases (such as db_bench, benchmark tool, stress/crash/unit tests) use a...
Added a few tests to start the discussion. Have not debugged them yet, but it appears like extra keys are thought to exist.
This PR replaces the code in the Options::Dump method with code using the Options infrastructure. A new "SerializePrintableOptions" method was added to the Configurable class which allows the values not...