Stop operating on DB in a stress test background thread

[riversand963]

Summary:
Stress test background threads do not coordinate with test worker
threads for db reopen in the middle of a test run, thus accessing db
obj in a stress test bg thread can race with test workers. Remove the
TimestampedSnapshotThread.

Test Plan:

./db_stress --acquire_snapshot_one_in=0 --adaptive_readahead=0 --allow_concurrent_memtable_write=1 \
--allow_data_in_errors=True --async_io=0 --avoid_flush_during_recovery=0 --avoid_unnecessary_blocking_io=1 \
--backup_max_size=104857600 --backup_one_in=100000 --batch_protection_bytes_per_key=8 \
--block_size=16384 --bloom_bits=7.580319535285394 --bottommost_compression_type=disable \
--bytes_per_sync=262144 --cache_index_and_filter_blocks=0 --cache_size=8388608 --cache_type=lru_cache \
--charge_compression_dictionary_building_buffer=1 --charge_file_metadata=0 --charge_filter_construction=1 \
--charge_table_reader=0 --checkpoint_one_in=0 --checksum_type=kxxHash64 --clear_column_family_one_in=0 \
--compact_files_one_in=1000000 --compact_range_one_in=0 --compaction_pri=1 --compaction_ttl=0 \
--compression_max_dict_buffer_bytes=0 --compression_max_dict_bytes=0 --compression_parallel_threads=1 \
--compression_type=xpress --compression_use_zstd_dict_trainer=1 --compression_zstd_max_train_bytes=0 \
--continuous_verification_interval=0 --create_timestamped_snapshot_one_in=20 --data_block_index_type=0 \
--db=/dev/shm/rocksdb/ --db_write_buffer_size=0 --delpercent=5 --delrangepercent=0 --destroy_db_initially=1 \
--detect_filter_construct_corruption=0 --disable_wal=0 --enable_compaction_filter=1 --enable_pipelined_write=0 \
--fail_if_options_file_error=1 --file_checksum_impl=xxh64 --flush_one_in=1000000 --format_version=2 \
--get_current_wal_file_one_in=0 --get_live_files_one_in=1000000 --get_property_one_in=1000000 \
--get_sorted_wal_files_one_in=0 --index_block_restart_interval=11 --index_type=0 --ingest_external_file_one_in=0 \
--iterpercent=0 --key_len_percent_dist=1,30,69 --level_compaction_dynamic_level_bytes=True \
--log2_keys_per_lock=10 --long_running_snapshots=0 --mark_for_compaction_one_file_in=10 \
--max_background_compactions=20 --max_bytes_for_level_base=10485760 --max_key=25000000 \
--max_key_len=3 --max_manifest_file_size=1073741824 --max_write_batch_group_size_bytes=64 \
--max_write_buffer_number=3 --max_write_buffer_size_to_maintain=0 --memtable_prefix_bloom_size_ratio=0.5 \
--memtable_whole_key_filtering=1 --memtablerep=skip_list --mmap_read=0 --mock_direct_io=True \
--nooverwritepercent=1 --open_files=500000 --open_metadata_write_fault_one_in=0 \
--open_read_fault_one_in=0 --open_write_fault_one_in=0 --ops_per_thread=20000 \
--optimize_filters_for_memory=1 --paranoid_file_checks=1 --partition_filters=0 --partition_pinning=2 \
--pause_background_one_in=1000000 --periodic_compaction_seconds=0 --prefix_size=1 \
--prefixpercent=5 --prepopulate_block_cache=0 --progress_reports=0 --read_fault_one_in=1000 \
--readpercent=55 --recycle_log_file_num=0 --reopen=100 --ribbon_starting_level=8 \
--secondary_cache_fault_one_in=0 --secondary_cache_uri= --snapshot_hold_ops=100000 \
--sst_file_manager_bytes_per_sec=104857600 --sst_file_manager_bytes_per_truncate=0 \
--subcompactions=3 --sync=0 --sync_fault_injection=0 --target_file_size_base=2097152 \
--target_file_size_multiplier=2 --test_batches_snapshots=0 --top_level_index_pinning=1 \
--txn_write_policy=0 --unordered_write=0 --unpartitioned_pinning=0 \
--use_direct_io_for_flush_and_compaction=0 --use_direct_reads=1 --use_full_merge_v1=1 \
--use_merge=1 --use_multiget=0 --use_txn=1 --user_timestamp_size=0 --value_size_mult=32 \
--verify_checksum=1 --verify_checksum_one_in=1000000 --verify_db_one_in=100000 \
--verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=0 --wal_compression=none \
--write_buffer_size=4194304 --write_dbid_to_manifest=0 --writepercent=35

make crash_test_with_txn
make crash_test_with_multiops_wc_txn

[facebook-github-bot]

@riversand963 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

[jay-zhuang]

LGTM.

Any plan to add it back?

[riversand963]

LGTM.

Any plan to add it back?

Thanks @jay-zhuang for the review!
These threads live beyond the life-cycle of databases, thus I doubt whether we need to do it.
It's possible if we can include these threads in the voting process to determine if database should be reopened. Currently, only the worker threads can vote.

[facebook-github-bot]

@riversand963 has updated the pull request. You must reimport the pull request before landing.

[facebook-github-bot]

@riversand963 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.