dolysis
/
rocksdb
mirror of https://github.com/facebook/rocksdb.git
3
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Sanitize checkpoint_one_in and lock_wal_one_in db_stress params (#13068) 

Summary:
Checkpoint creation skips flushing the memtable, even if explicitly requested, when the WAL is locked. This can happen if the user calls `LockWAL()`. In this case, db_stress checkpoint verification fails as the checkpoint will not contain keys present in the primary DB's memtable. Sanitize `checkpoint_one_in` and `lock_wal_one_in` so they're mutually exclusive.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/13068

Reviewed By: hx235

Differential Revision: D64353998

Pulled By: anand1976

fbshipit-source-id: 7c93563347f033b6008a47a7d71471e59747e143
This commit is contained in:
anand76 2024-10-14 21:11:37 -07:00 committed by Facebook GitHub Bot
parent dd76862b00
commit cbebbad7d9
1 changed files with 6 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
tools/db_crashtest.py
View File

@ -47,7 +47,7 @@ default_params = {
"charge_filter_construction": lambda: random.choice([0, 1]),
"charge_table_reader": lambda: random.choice([0, 1]),
"charge_file_metadata": lambda: random.choice([0, 1]),
"checkpoint_one_in": lambda: random.choice([10000, 1000000]),
"checkpoint_one_in": lambda: random.choice([0, 0, 10000, 1000000]),
"compression_type": lambda: random.choice(
["none", "snappy", "zlib", "lz4", "lz4hc", "xpress", "zstd"]
),
@ -965,6 +965,10 @@ def finalize_and_sanitize(src_params):
):
# At least one must be true
dest_params["write_dbid_to_manifest"] = 1
# Checkpoint creation skips flush if the WAL is locked, so enabling lock_wal_one_in
# can cause checkpoint verification to fail. So make the two mutually exclusive.
if dest_params.get("checkpoint_one_in") != 0:
dest_params["lock_wal_one_in"] = 0
return dest_params