Andrew Kryczka

> The second approach sounds cleaner since it does not require consecutive sequence numbers in consecutive WALs. For the first approach, a dummy record with a new type at the...

`SetOptions()` sanitization is a longstanding issue: https://github.com/facebook/rocksdb/blob/fc40165614e3aecc1475d61f895f508b28247c0d/include/rocksdb/db.h#L1452-L1454

Sorry you are right that negative compaction scores can happen for L0 when L0 compaction is disabled (-1). That just means compaction won't happen. You can change them to report...

Could you please copy/paste the compiler output instead of providing a screenshot? Otherwise, important parts of the issue will not be searchable.

What allocator are you using? We use jemalloc. If you are using jemalloc, can you set `dump_malloc_stats`?

> looks like I can set an allocator for block cache? My application that has rocksdb bundled in it is using jemalloc, but I don't pass a memory allocator when...

> a feature request that might be helpful is to include in the LOG the allocator being used by rocksdb to make sure that users don't think they are using...

#7428 I guess. We could clarify on the wiki that it shows a suggested shutdown procedure, not a required one. Otherwise people who aren't following the procedure may think something...

If `mmap()` and storing the data uncompressed are both OK then plain table may be of use - https://github.com/facebook/rocksdb/wiki/PlainTable-Format, https://github.com/facebook/rocksdb/wiki/RocksDB-In-Memory-Workload-Performance-Benchmarks Otherwise you'd use block-based table and configure a large enough...

Yes. Just make sure the `prefix_extractor` you are setting reflects what kind of range scans you will be doing (with plain table, range scan needs to be within a key...