Summary:
Separate flag for enabling option from flag for enabling dedicated atomic stress test. I have found setting the former without setting the latter can detect different problems.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4781
Differential Revision: D13463211
Pulled By: ajkr
fbshipit-source-id: 054f777885b2dc7d5ea99faafa21d6537eee45fd
Summary:
The default for index_block_restart_interval is 1 but some use 16 in production. The patch extends crash test to test both values.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4383
Differential Revision: D9887304
Pulled By: maysamyabandeh
fbshipit-source-id: a8d00fea974a79ad563f9f4d9d7b069e9f746a8f
Summary:
- Add `--compression_max_dict_bytes` and `--compression_zstd_max_train_bytes` flags to stress test
- Randomly enable/disable the above flags in crash test
- Set `--compression_type=zstd` in FB-specific crash test runs
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4234
Differential Revision: D9187207
Pulled By: ajkr
fbshipit-source-id: 8d78cf8d8e1165f2cd1c32e069b73726b5bc1fd2
Summary:
db_stress doesn't cover upper or lower bound in iterators. Try to cover it by randomly assigning a random one. Also in prefix scan tests, with 50% of the chance, set next prefix as the upper bound.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4162
Differential Revision: D8953507
Pulled By: siying
fbshipit-source-id: f0f04e9cb6c07cbebbb82b892ca23e0daeea708b
Summary:
Add the `backup_one_in` and `checkpoint_one_in` options to periodically trigger backups and checkpoints. The directory names contain thread ID to avoid clashing with parallel backups/checkpoints. Enable checkpoint in crash test so our CI runs will use it. Didn't enable backup in crash test since it copies all the files which is too slow.
Closes https://github.com/facebook/rocksdb/pull/4005
Differential Revision: D8472275
Pulled By: ajkr
fbshipit-source-id: ff91bdc37caac4ffd97aea8df96b3983313ac1d5
Summary:
Fixed bug where `db_stress` output a line with a warning followed by a line with an error, and `db_crashtest.py` considered that a success. For example:
```
WARNING: prefix_size is non-zero but memtablerep != prefix_hash
open error: Corruption: SST file is ahead of WALs
```
Closes https://github.com/facebook/rocksdb/pull/4006
Differential Revision: D8473463
Pulled By: ajkr
fbshipit-source-id: 60461bdd7491d9d26c63f7d4ee522a0f88ba3de7
Summary:
format_version=3 changes the format of SST index. This is however not being tested currently since tests only work with the default format_version which is currently 2. The patch extends the most related tests to also test for format_version=3.
Closes https://github.com/facebook/rocksdb/pull/3942
Differential Revision: D8238413
Pulled By: maysamyabandeh
fbshipit-source-id: 915725f55753dd8e9188e802bf471c23645ad035
Summary:
We need to keep the DB directory around since the direct IO check in "db_crashtest.py" relies on it existing. This PR fixes an issue where it was removed after each stress test run during the second half of whitebox crash testing.
Closes https://github.com/facebook/rocksdb/pull/3946
Differential Revision: D8247998
Pulled By: ajkr
fbshipit-source-id: 4e7cffbdab9b40df125e7842d0d59916e76261d3
Summary:
Previously `db_stress` attempted to configure direct I/O dynamically in `SetOptions()` which had multiple problems (ummm must've never been tested):
- It's a DB option so SetDBOptions should've been called instead
- It's not a dynamic option so even SetDBOptions would fail
- It required enabling SyncPoint to mask O_DIRECT since it had no way to detect whether the DB directory was in tmpfs or not. This required locking that consumed ~80% of db_stress CPU.
In this PR I delete the broken dynamic config and instead configure it statically, only enabling it if the DB directory truly supports O_DIRECT.
Closes https://github.com/facebook/rocksdb/pull/3939
Differential Revision: D8238120
Pulled By: ajkr
fbshipit-source-id: 60bb2deebe6c9b54a3f788079261715b4a229279
Summary:
- Any options unknown to `db_crashtest.py` are now passed directly to `db_stress`. This way, we won't need to update `db_crashtest.py` every time `db_stress` gets a new option.
- Remove `db_crashtest.py` redundant arguments where the value is the same as `db_stress`'s default
- Remove `db_crashtest.py` redundant arguments where the value is the same in a previously applied options map. For example, default_params are always applied before whitebox_default_params, so if they require the same value for an argument, that value only needs to be provided in default_params.
- Made the simple option maps applied in addition to the regular option maps. Previously they were exclusive which led to lots of duplication
Closes https://github.com/facebook/rocksdb/pull/3809
Differential Revision: D7885779
Pulled By: ajkr
fbshipit-source-id: 3a3243b55724d6d5bff36e939b582b9b62c538a8
Summary:
- Original commit: a4fb1f8c04
- Revert commit (we reverted as a quick fix to get crash tests passing): 6afe22db2e
This PR includes the contents of the original commit plus two bug fixes, which are:
- In whitebox crash test, only set `--expected_values_path` for `db_stress` runs in the first half of the crash test's duration. In the second half, a fresh DB is created for each `db_stress` run, so we cannot maintain expected state across `db_stress` runs.
- Made `Exists()` return true for `UNKNOWN_SENTINEL` values. I previously had an assert in `Exists()` that value was not `UNKNOWN_SENTINEL`. But it is possible for post-crash-recovery expected values to be `UNKNOWN_SENTINEL` (i.e., if the crash happens in the middle of an update), in which case this assertion would be tripped. The effect of returning true in this case is there may be cases where a `SingleDelete` deletes no data. But if we had returned false, the effect would be calling `SingleDelete` on a key with multiple older versions, which is not supported.
Closes https://github.com/facebook/rocksdb/pull/3793
Differential Revision: D7811671
Pulled By: ajkr
fbshipit-source-id: 67e0295bfb1695ff9674837f2e05bb29c50efc30
Summary:
crash-recovery verification is failing in the whitebox testing, which may or may not be a valid correctness issue -- need more time to investigate. In the meantime, reverting so we don't mask other failures.
Closes https://github.com/facebook/rocksdb/pull/3786
Differential Revision: D7794516
Pulled By: ajkr
fbshipit-source-id: 28ccdfdb9ec9b3b0fb08c15cbf9d2e282201ff33
Summary:
- When options file is provided to db_stress, take supported options from the file instead of from flags
- Call `BuildOptionsTable` after `Open` so it can use `options_` once it has been populated either from flags or from file
- Allow options filename to be passed via `db_crashtest.py`
Closes https://github.com/facebook/rocksdb/pull/3768
Differential Revision: D7755331
Pulled By: ajkr
fbshipit-source-id: 5205cc5deb0d74d677b9832174153812bab9a60a
Summary:
Previously, our `db_stress` tool held the expected state of the DB in-memory, so after crash-recovery, there was no way to verify data correctness. This PR adds an option, `--expected_values_file`, which specifies a file holding the expected values.
In black-box testing, the `db_stress` process can be killed arbitrarily, so updates to the `--expected_values_file` must be atomic. We achieve this by `mmap`ing the file and relying on `std::atomic<uint32_t>` for atomicity. Actually this doesn't provide a total guarantee on what we want as `std::atomic<uint32_t>` could, in theory, be translated into multiple stores surrounded by a mutex. We can verify our assumption by looking at `std::atomic::is_always_lock_free`.
For the `mmap`'d file, we didn't have an existing way to expose its contents as a raw memory buffer. This PR adds it in the `Env::NewMemoryMappedFileBuffer` function, and `MemoryMappedFileBuffer` class.
`db_crashtest.py` is updated to use an expected values file for black-box testing. On the first iteration (when the DB is created), an empty file is provided as `db_stress` will populate it when it runs. On subsequent iterations, that same filename is provided so `db_stress` can check the data is as expected on startup.
Closes https://github.com/facebook/rocksdb/pull/3629
Differential Revision: D7463144
Pulled By: ajkr
fbshipit-source-id: c8f3e82c93e045a90055e2468316be155633bd8b
Summary:
Previously `python tools/db_crashtest.py blackbox` would do no useful work as the crash interval (two minutes) was shorter than the preparation phase. The preparation phase is slow because of the ridiculously inefficient way it computes which keys should not be overwritten. It was doing this for 60M keys since default values were `FLAGS_nooverwritepercent == 60` and `FLAGS_max_key == 100000000`.
Move the "nooverwritepercent" override from whitebox-specific to the general options so it also applies to blackbox test runs. Now preparation phase takes a few seconds.
Closes https://github.com/facebook/rocksdb/pull/3671
Differential Revision: D7457732
Pulled By: ajkr
fbshipit-source-id: 601f4461a6a7e49e50449dcf15aebc9b8a98d6f0
Summary: Grandfather in super old lint issues to make a clean slate for moving forward that allows us to have stronger enforcement on new issues.
Reviewed By: yiwu-arbug
Differential Revision: D6821806
fbshipit-source-id: 22797d31ec58e9eb0255d3b66fedfcfcb0dc127c
Summary:
Add options to `db_stress` (correctness testing tool) to randomly acquire snapshot and release it after some period of time. It's useful for correctness testing of #3009, as well as other parts of compaction that behave differently depending on which snapshots are held.
Closes https://github.com/facebook/rocksdb/pull/3038
Differential Revision: D6086501
Pulled By: ajkr
fbshipit-source-id: 3ec0d8666c78ac507f1f808887c4ff759ba9b865
Summary:
TSAN shows error when we grab too many locks at the same time. In TSAN crash test, make one shard key cover 2^22 keys so that no many keys will be hold at the same time.
Closes https://github.com/facebook/rocksdb/pull/2719
Differential Revision: D5609035
Pulled By: siying
fbshipit-source-id: 930e5d63fff92dbc193dc154c4c615efbdf06c6a
Summary:
filter the warning out and only print it once.
Closes https://github.com/facebook/rocksdb/pull/2137
Differential Revision: D4870925
Pulled By: lightmark
fbshipit-source-id: 91b363ce7f70bce88b0780337f408fc4649139b8
Summary:
Remove disableDataSync, and another similarly named disable_data_sync options.
This is being done to simplify options, and also because the performance gains of this feature can be achieved by other methods.
Closes https://github.com/facebook/rocksdb/pull/1859
Differential Revision: D4541292
Pulled By: sagar0
fbshipit-source-id: 5b3a6ca
Summary:
Clock-based cache implemenetation aim to have better concurreny than
default LRU cache. See inline comments for implementation details.
Test Plan:
Update cache_test to run on both LRUCache and ClockCache. Adding some
new tests to catch some of the bugs that I fixed while implementing the
cache.
Reviewers: kradhakrishnan, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D61647
Summary:
Tsan crash white-box test was disabled because it never ends. Re-enable it with reduced killing odds.
Add a parameter in crash test script to allow we pass it through an environment variable.
Test Plan: Run it manually.
Reviewers: IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D61053
Summary: filter_deletes option was removed, remove it from crash_test to fix it
Test Plan: make crash_test
Reviewers: yhchiang, andrewkr, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D59901
Summary: Default crash test uses prefix hash memtable, which is not compatible to concurrent memtable. Allow prefix test run with skip list and use skip list memtable when concurrent insert is used.
Test Plan: Run "python -u tools/db_crashtest.py whitebox" and watch sometimes skip list is used.
Reviewers: anthony, yhchiang, kradhakrishnan, andrewkr, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D53907
Summary: If users turn on concurrent insert but the memtable doesn't support it, they might see unexcepted crash. Fix it by explicitly fail.
Test Plan:
Run different setting of stress_test and make sure it fails correctly.
Will add a unit test too.
Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, andrewkr, ngbronson
Reviewed By: ngbronson
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D53895
Summary: Add an option of --allow_concurrent_memtable_write in stress test and cover it in crash test
Test Plan: Run crash test and make sure three combinations of the two options show up randomly.
Reviewers: IslamAbdelRahman, yhchiang, andrewkr, anthony, kradhakrishnan
Reviewed By: kradhakrishnan
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D53811
Summary:
In crash test, when coming to each kill point, we start a random class using seed as current second. With this approach, for every second, the random number used is the same. However, in each second, there are multiple kill points with different frequency. It makes it hard to reason about chance of kill point to trigger. With this commit, we use thread local random seed to generate the random number, so that it will take different values per second, hoping it makes chances of killing much easier to reason about.
Also significantly reduce the kill odd to make sure time before kiling is similar as before.
Test Plan: Run white box crash test and see the killing happens as expected and the run time time before killing reasonable.
Reviewers: kradhakrishnan, IslamAbdelRahman, rven, yhchiang, andrewkr, anthony
Reviewed By: anthony
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D52971
Summary: crash_test still has a very low chance to hit some crash point. Have another mode for covering them more likely.
Test Plan: Run crash_test and see db_stress is called with expected prameters.
Reviewers: kradhakrishnan, igor, anthony, rven, IslamAbdelRahman, yhchiang
Reviewed By: yhchiang
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D49473
Summary:
in tools/db_crashtest.py, cmd_params['db'] by default is a lambda expression, not the actual db_name.
fix by get the db_name before passing it to gen_cmd.
Test Plan: run `make crashtest`
Reviewers: sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D49119
Summary:
merge tools/db_crashtest2.py into tools/db_crashtest.py
python tools/db_crashtest.py -h # show help message, ALL parameters can be overwrite by arguments
Example usages:
python tools/db_crashtest.py blackbox # run blackbox with default parameters
python tools/db_crashtest.py blackbox --simple
python tools/db_crashtest.py whitebox # run whitebox with default parameters
python tools/db_crashtest.py whitebox --simple
all default parameters are identical to previous version.
Test Plan: `make crash_test` and make sure it can run with expected parameters pased to db_stress.
Reviewers: igor, rven, anthony, IslamAbdelRahman, yhchiang, sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D48567
Summary: Mode of data sync off is a much less used than the case of data sync on. Crash test should cover the more common case than a corner case. So turn data sync on in crash tests.
Test Plan: Run crash test and make sure it can run with expected parameters pased to db_stres.
Reviewers: IslamAbdelRahman, anthony, igor, rven, kradhakrishnan, yhchiang
Reviewed By: yhchiang
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D48729
Summary: In a recent change, crash_test can put data under TEST_TMPDIR. However, the directory is not cleaned before running the test, which may cause unexpected results. Clean it.
Test Plan: Run white and black box crash test against non-existing, or non-empty but not compactible DBs, and make sure it works as expected.
Reviewers: kradhakrishnan, rven, yhchiang, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D43515
Summary: Currently crash_test only puts data under /tmp. It is less flexible if we want to cover different file systems or media. Make crash_test to appreciate TEST_TMPDIR so that users can run it against another file system.
Test Plan: Run blackbox_crash_test and whitebox_crash_test with or without TEST_TMPDIR set and make sure DBs are put in the right place
Reviewers: kradhakrishnan, yhchiang, rven, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D43509
Summary:
crash_test now only runs complicated options, multiple column families, prefix hash, frequently changing options, many compaction threads, etc. These options are good to cover new features but we loss coverage in most common use cases. Furthermore, by running only for multiple column families, we are not able to create LSM trees that are large enough to cover some stress cases.
Make half of crash_test runs the simply tests: single column family, default mem table, one compaction thread, no change options.
Test Plan: Run crash_test
Reviewers: rven, yhchiang, IslamAbdelRahman, kradhakrishnan
Reviewed By: kradhakrishnan
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D43461
Summary: Allow SetOptions() during db_stress test
Test Plan: make crash_test
Reviewers: sdong, yhchiang, rven, igor
Reviewed By: igor
Subscribers: leveldb
Differential Revision: https://reviews.facebook.net/D25497
Summary:
As discussed in our internal group, we don't get much use of seek compaction at the moment, while it's making code more complicated and slower in some cases.
This diff removes seek compaction and (hopefully) all code that was introduced to support seek compaction.
There is one test case that relied on didIO information. I'll try to find another way to implement it.
Test Plan: make check
Reviewers: sdong, haobo, yhchiang, ljin, dhruba
Reviewed By: ljin
Subscribers: leveldb
Differential Revision: https://reviews.facebook.net/D19161
Summary:
We should clean up after ourselves. Last crashtest failed because the disk on the cibuild machine was full.
Also, db_crashtest is supposed to run the test on the same database in every iteration. This isn't the case currently, because every run creates a new tmp directory. It's fixed with this diff.
Test Plan: ran it
Reviewers: ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17073
Summary:
Hash skip list has issues, causing db_stress to fail badly.
For now, switching back to skip_list by default before we figure out root cause.
Test Plan: db_stress is happy(ier)
Reviewers: ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16845
Summary: Syncing in stress test makes it run much much much slower. It also doesn't add much value IMO.
Test Plan: no
Reviewers: ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16839
Summary:
*) fixed the comment
*) constant 1 was not casted to 64-bit, which (I think) might cause overflow if we shift it too much
*) default prefix size to be 7, like it was before
Test Plan: compiled
Reviewers: ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16827
Summary: Fix the db_stress test, let is run with HashSkipList for real
Test Plan:
python tools/db_crashtest.py
python tools/db_crashtest2.py
Reviewers: igor, haobo
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16773
Summary:
Why don't we automatically reopen DB when running crash test (running in our nightly build)? If I understand correctly, crashtest is manually reopenning the DB, but then the DB does not check its consistency when you kill db_stress process and then re-run it again.
Does this make sense?
Test Plan: not reall
Reviewers: dhruba, haobo, emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16167
Summary: Adding in the iterpercent flag to tests.
Test Plan: make crash_test
Reviewers: emayanke
Reviewed By: emayanke
Differential Revision: https://reviews.facebook.net/D13035
Summary:
Most code change in this diff is code cleanup/rewrite. The logic changes include:
(1) add universal compaction to db_crashtest2.py
(2) randomly set --test_batches_snapshots to be 0 or 1 in db_crashtest2.py. Old codes always use 1.
(3) use different tmp directory as db directory in different runs. I saw some intermittent errors in my local tests. Use of different tmp directory seems to be able to solve the issue.
Test Plan: Have run "make crashtest" for multiple times. Also run "make all check"
Reviewers: emayanke, dhruba, haobo
Reviewed By: emayanke
Differential Revision: https://reviews.facebook.net/D12369
Summary: I changed the db_stress configs, but forgot to update the scripts using the old configs.
Test Plan: 'make blackbox_crash_test' and 'make whitebox_crash_test' start running normally now (I haven't run them til the end, though).
Reviewers: vamsi
Reviewed By: vamsi
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D12303
Summary: filter_deletes options introduced in db_stress makes it drop Deletes on key if KeyMayExist(key) returns false on the key. code change was simple and tested so not wasting reviewer's time.
Test Plan: maek crash_test; python tools/db_crashtest[1|2].py
CC: dhruba, vamsi
Differential Revision: https://reviews.facebook.net/D11769
Summary: As title, now that db_stress supports --map_read properly
Test Plan: make crash_test
Reviewers: vamsi, emayanke, dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11391
Summary: These extra options caught some bugs. Will be run via Jenkins now with the crash_test
Test Plan: ./make crashtest
Reviewers: dhruba, vamsi
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11151
Summary: To know which options the crashtest was run with. Also changed print to sys.stdout.write which is more standard.
Test Plan: python tools/db_crashtest.py
Reviewers: vamsi, akushner, dhruba
Reviewed By: akushner
Differential Revision: https://reviews.facebook.net/D10119
Summary: make crash_test will now invoke the crash_test. Also some cleanup in the db_crashtest.py file
Test Plan: make crash_test
Reviewers: akushner, vamsi, sheki, dhruba
Reviewed By: vamsi
Differential Revision: https://reviews.facebook.net/D9987
Summary:
When I run db_crashtest, I am seeing lot of warnings that say db_stress completed
before it was killed. To fix that I made ops per thread a very large value so that it keeps
running until it is killed.
I also set #reopens to 0. Since we are killing the process anyway, the 'simulated crash'
that happens during reopen may not add additional value.
I usually see 10-25K ops happening before the kill. So I increased max_key from 100 to
1000 so that we use more distinct keys.
Test Plan:
Ran a few times.
Revert Plan: OK
Task ID: #
Reviewers: emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9909
Summary: The script runs and kills the stress test periodically. Default values have been used in the script now. Should I make this a part of the Makefile or automated rocksdb build? The values can be easily changed in the script right now, but should I add some support for variable values or input to the script? I believe the script achieves its objective of unsafe crashes and reopening to expect sanity in the database.
Test Plan: python tools/db_crashtest.py
Reviewers: dhruba, vamsi, MarkCallaghan
Reviewed By: vamsi
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9369
