Commit Graph

1275 Commits

Author SHA1 Message Date
Igor Canadi
fcd5c5e828 ComputeCompactionScore in CompactionPicker
Summary:
As it turns out, we need the call to ComputeCompactionScore (previously: Finalize) in CompactionPicker.

The issue caused a deadlock in db_stress: http://ci-builds.fb.com/job/rocksdb_crashtest/290/console

The last two lines before a deadlock were:
2014/03/18-22:43:41.481029 7facafbee700 (Original Log Time 2014/03/18-22:43:41.480989) Compaction nothing to do
2014/03/18-22:43:41.481041 7faccf7fc700 wait for fewer level0 files...

"Compaction nothing to do" and other thread waiting for fewer level0 files. Hm hm.

I moved the pre-sorting to SaveTo, which should fix both the original and the new issue.

Test Plan: make check for now, will run db_stress in jenkins

Reviewers: dhruba, haobo, sdong

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D17037
2014-03-19 16:52:26 -07:00
Kai Liu
69f6cf431d Fix two bugs in talbe format
Summary:
Previous code had two bugs:

* didn't initialize the table_magic_number_ explicitly -- as a
  result a random junk number is stored for table_magic_number_, making
  HasInitializedMagicNumber() always return true.
* if condition is inconrrect in set_table_magic_number(), and the return value is not checked.
  I replace if-else by a stronger requirement enforced by assert().

Test Plan:
Previous sst_dump failed to work.
After the fix, things back to normal.

Reviewers: yhchiang

CC: haobo, sdong, igor, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D17055
2014-03-19 16:18:33 -07:00
Igor Canadi
e493f2f54e Don't compact with zero input files
Summary:
We have an issue with internal service trying to run compaction with zero input files:
2014/02/07-02:26:58.386531 7f79117ec700 Compaction start summary: Base version 1420 Base level 3, seek compaction:0, inputs:[ϛ~^Qy^?],[]
2014/02/07-02:26:58.386539 7f79117ec700 Compacted 0@3 + 0@4 files => 0 bytes

There are two issues:
* inputsummary is printing out junk
* it's constantly retrying (since I guess madeProgress is true), so it prints out a lot of data in the LOG file (40GB in one day).

I read through the Level compaction picker and added some failure condition if input[0] is empty. I think PickCompaction() should not return compaction with zero input files with this change. I'm not confident enough to add an assertion though :)

Test Plan: make check

Reviewers: dhruba, haobo, sdong, kailiu

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16005
2014-03-19 16:01:25 -07:00
Igor Canadi
1ad0c2f9db add tags to gitignore 2014-03-19 15:40:33 -07:00
Igor Canadi
22507aff6c Fix compile issue in Mac OS
Summary:
Compile issues are:
* Unused variable env_
* Unused fallocate_with_keep_size_

Test Plan: compiles

Reviewers: dhruba, haobo, sdong

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D17043
2014-03-19 15:40:12 -07:00
Lei Jin
6dc940d4c9 avoid shared_ptr assignment in Version::Get()
Summary:
This is a 500ns operation while the whole Get() call takes only a few
micro!

Test Plan: ran db_bench, for a DB with 50M keys, QPS jumps from 5.2M/s to 7.2M/s

Reviewers: haobo, igor, dhruba

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D17007
2014-03-19 10:54:32 -07:00
sdong
71e6a34271 Add a DB property to indicate number of background errors encountered
Summary: Add a property to calculate number of background errors encountered to help users build their monitoring

Test Plan: Add a unit test. make all check

Reviewers: haobo, igor, dhruba

Reviewed By: igor

CC: ljin, nkg-, yhchiang, leveldb

Differential Revision: https://reviews.facebook.net/D16959
2014-03-18 14:28:30 -07:00
Kai Liu
1ec72b37b1 Several easy-to-add properties related to compaction and flushes
Summary: To partly address the request @nkg- raised, add three easy-to-add properties to compactions and flushes.

Test Plan: run unit tests and add a new unit test to cover new properties.

Reviewers: haobo, dhruba

Reviewed By: dhruba

CC: nkg-, leveldb

Differential Revision: https://reviews.facebook.net/D13677
2014-03-18 14:00:09 -07:00
Igor Canadi
758fa8c359 Don't Finalize in CompactionPicker
Summary:
Finalize re-sorts (read: mutates) the files_ in Version* and it is called by CompactionPicker during normal runtime. At the same time, this same Version* lives in the SuperVersion* and is accessed without the mutex in GetImpl() code path.

Mutating the files_ in one thread and reading the same files_ in another thread is a bad idea. It caused this issue: http://ci-builds.fb.com/job/rocksdb_crashtest/285/console

Long-term, we need to be more careful with method contracts and clearly document what state can be mutated when. Now that we are much faster because we don't lock in GetImpl(), we keep running into data races that were not a problem before when we were slower. db_stress has been very helpful in detecting those.

Short-term, I removed Finalize() from CompactionPicker.

Note: I believe this is an issue in current 2.7 version running in production.

Test Plan:
make check
Will also run db_stress to see if issue is gone

Reviewers: sdong, ljin, dhruba, haobo

Reviewed By: sdong

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16983
2014-03-18 13:59:59 -07:00
Lei Jin
63cef90078 disable the log_number check in Recover()
Summary:
There is a chance that an old MANIFEST is corrupted in 2.7 but just not noticed.
This check would fail them. Change it to log instead of returning a
Corruption status.

Test Plan: make

Reviewers: haobo, igor

Reviewed By: igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16923
2014-03-18 12:46:29 -07:00
Yueh-Hsuan Chiang
7624f43e0a Fixed a typo in INSTALL.md
Summary: Replace "RocskDB" by "RocksDB" in INSTALL.md

Test Plan: No code change.

Reviewers: ljin, igor

Reviewed By: ljin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16977
2014-03-18 12:05:23 -07:00
Igor Canadi
f26cb0f093 Optimize fallocation
Summary:
Based on my recent findings (posted in our internal group), if we use fallocate without KEEP_SIZE flag, we get superior performance of fdatasync() in append-only workloads.

This diff provides an option for user to not use KEEP_SIZE flag, thus optimizing his sync performance by up to 2x-3x.

At one point we also just called posix_fallocate instead of fallocate, which isn't very fast: http://code.woboq.org/userspace/glibc/sysdeps/posix/posix_fallocate.c.html (tl;dr it manually writes out zero bytes to allocate storage). This diff also fixes that, by first calling fallocate and then posix_fallocate if fallocate is not supported.

Test Plan: make check

Reviewers: dhruba, sdong, haobo, ljin

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16761
2014-03-17 21:52:14 -07:00
Igor Canadi
ae25742af9 Fix race condition in manifest roll
Summary:
When the manifest is getting rolled the following happens:
1) manifest_file_number_ is assigned to a new manifest number (even though the old one is still current)
2) mutex is unlocked
3) SetCurrentFile() creates temporary file manifest_file_number_.dbtmp
4) SetCurrentFile() renames manifest_file_number_.dbtmp to CURRENT
5) mutex is locked

If FindObsoleteFiles happens between (3) and (4) it will:
1) Delete manifest_file_number_.dbtmp (because it's not in pending_outputs_)
2) Delete old manifest (because the manifest_file_number_ already points to a new one)

I introduce the concept of prev_manifest_file_number_ that will avoid the race condition.

However, we should discuss the future of MANIFEST file rolling. We found some race conditions with it last week and who knows how many more are there. Nobody is using it in production because we don't trust the implementation. Should we even support it?

Test Plan: make check

Reviewers: ljin, dhruba, haobo, sdong

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16929
2014-03-17 21:50:15 -07:00
Igor Canadi
5601bc4619 Check starts_with(prefix) in MultiPrefixIterate
Summary: We switched to prefix_seek method of seeking. This means that anytime we check Valid(), we also need to check starts_with(prefix)

Test Plan: ran db_stress

Reviewers: ljin

Reviewed By: ljin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16953
2014-03-17 17:02:34 -07:00
Igor Canadi
9caeff516e keep_log_files option in BackupableDB
Summary:
Added an option to BackupableDB implementation that allows users to persist in-memory databases. When the restore happens with keep_log_files = true, it will
*) Not delete existing log files in wal_dir
*) Move log files from archive directory to wal_dir, so that DB can replay them if necessary

Test Plan: Added an unit test

Reviewers: dhruba, ljin

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16941
2014-03-17 15:39:23 -07:00
Yueh-Hsuan Chiang
a5fafd4f46 Correct the logic of MemTable::ShouldFlushNow().
Summary:
Memtable will now be forced to flush if the one of the following
conditions is met:
1. Already allocated more than write_buffer_size + 60% arena block size.
   (the overflowing condition)
2. Unable to safely allocate one more arena block without hitting the
   overflowing condition AND the unused allocated memory < 25% arena
   block size.

Test Plan: make all check

Reviewers: sdong, haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16893
2014-03-17 12:20:11 -07:00
Igor Canadi
9b8a2b52d4 No prefix iterator in db_stress
Summary: We're trying to deprecate prefix iterators, so no need to test them in db_stress

Test Plan: ran it

Reviewers: ljin

Reviewed By: ljin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16917
2014-03-17 10:02:46 -07:00
sdong
c61c9830d4 Fix a bug that Prev() can hang.
Summary: Prev() now can hang when there is a key with more than max_skipped number of appearance internally but all of them are newer than the sequence ID to seek. Add unit tests to confirm the bug and fix it.

Test Plan: make all check

Reviewers: igor, haobo

Reviewed By: igor

CC: ljin, yhchiang, leveldb

Differential Revision: https://reviews.facebook.net/D16899
2014-03-17 10:00:41 -07:00
Igor Canadi
f9d0530213 Don't care about signed/unsigned compare
Summary:
We need to stop these:
https://github.com/facebook/rocksdb/pull/99
https://github.com/facebook/rocksdb/pull/83

Test Plan: no

Reviewers: dhruba, haobo, sdong, ljin, yhchiang

Reviewed By: ljin

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16905
2014-03-17 09:41:41 -07:00
Igor Canadi
30447b7251 Merge pull request #99 from caiosba/master
Make it compile on Debian/GCC 4.7
2014-03-17 08:51:35 -07:00
Lei Jin
0cf6c8f7ce fix: use the correct edit when comparing log_number
Summary:
In the last fix, I forgot to point to the writer when comparing edit,
which is apparently not correct.

Test Plan: still running make whitebox_crash_test

Reviewers: igor, haobo, igor2

Reviewed By: igor2

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16911
2014-03-15 23:30:43 -07:00
Lei Jin
453ec52ca1 journal log_number correctly in MANIFEST
Summary:
Here is what it can cause probelm:
There is one memtable flush and one compaction. Both call LogAndApply(). If both edits are applied in the same batch with flush edit first and the compaction edit followed. LogAndApplyHelper() will assign compaction edit current VersionSet's log number(which should be smaller than the log number from flush edit). It cause log_numbers in MANIFEST to be not monotonic increasing, which violates the assume Recover() makes. What is more is after comitting to MANIFEST file, log_number_ in VersionSet is updated to the log_number from the last edit, which is the compaction one. It ends up not updating the log_number.

Test Plan:
make whitebox_crash_test
got another assertion about iter->valid(), not sure if that is related
to this.

Reviewers: igor, haobo

Reviewed By: igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16875
2014-03-14 18:36:47 -07:00
Caio SBA
f234dfd8fb Breaking line 2014-03-14 23:56:58 +00:00
Caio SBA
b9c78d2db6 Make it compile on Debian/GCC 4.7 2014-03-14 22:44:35 +00:00
Igor Canadi
5948a6632d Merge pull request #97 from agchou/patch-1
Fix copyright year
2014-03-14 15:38:35 -07:00
Igor Canadi
2bad3cb0db Missing includes 2014-03-14 13:02:20 -07:00
Igor Canadi
56dce9bf8e unterminated conditional directive 2014-03-14 10:22:37 -07:00
Igor Canadi
f74659ac9f Fix another Mac OS warning 2014-03-14 10:13:39 -07:00
Igor Canadi
3c75cc15a9 Fix HashSkipList and HashLinkedList SIGSEGV
Summary:
Original Summary:
Yesterday, @ljin and I were debugging various db_stress issues. We suspected one of them happens when we concurrently call NewIterator without prefix_seek on HashSkipList. This test demonstrates it.

Update:
Arena is not thread-safe!! When creating a new full iterator, we *have* to create a new arena, otherwise we're doomed.

Test Plan: SIGSEGV and assertion-throwing test now works!

Reviewers: ljin, haobo, sdong

Reviewed By: sdong

CC: leveldb, ljin

Differential Revision: https://reviews.facebook.net/D16857
2014-03-14 10:02:04 -07:00
Igor Canadi
6c72079d77 Fix warning on Mac OS 2014-03-14 09:54:23 -07:00
sdong
5aa81f04fa Fix extra compaction tasks scheduled after D16767 in some cases
Summary:
With D16767, there is a case compaction tasks are scheduled infinitely:
(1) no flush thread is configured and more than 1 compaction threads
(2) a flush is going on by one compaction hread
(3) the state of SST files is in the state that versions_->current()->NeedsCompaction() will generate a false positive (return true actually there is no work to be done)
In that case, a infinite loop will be formed.

This patch would fix it.

Test Plan: make all check

Reviewers: haobo, igor, ljin

Reviewed By: igor

CC: dhruba, yhchiang, leveldb

Differential Revision: https://reviews.facebook.net/D16863
2014-03-13 13:06:08 -07:00
Igor Canadi
04a1035efe Revert "DB stress with normal skip list"
This reverts commit 86926d8c6a.
2014-03-12 22:21:13 -07:00
Lei Jin
02a2cb139b fix VerifyDb in StressTest
Summary:
this should fix the hash_skip_list issue, but I still see seqno
assertion failure in the last run. Will continue investigating and
address that in a different diff

Test Plan: make whitebox_crash_test

Reviewers: igor

Reviewed By: igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16851
2014-03-12 22:20:22 -07:00
Igor Canadi
86926d8c6a DB stress with normal skip list
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
2014-03-12 17:35:27 -07:00
Kai Liu
11da8bc5df A heuristic way to check if a memtable is full
Summary:
This is is based on https://reviews.facebook.net/D15027. It's not finished but I would like to give a prototype to avoid arena over-allocation while making better use of the already allocated memory blocks.

Instead of check approximate memtable size, we will take a deeper look at the arena, which incorporate essential idea that @sdong suggests: flush when arena has allocated its last and the last is "almost full"

Test Plan: N/A

Reviewers: haobo, sdong

Reviewed By: sdong

CC: leveldb, sdong

Differential Revision: https://reviews.facebook.net/D15051
2014-03-12 16:40:14 -07:00
Igor Canadi
7b7793e97a Don't sync in stress test
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
2014-03-12 15:12:09 -07:00
agchou
b694b16105 Fix copyright year
Fix outdated copyright year (update to 2014)

The copyright year was out of date. Copyright notices must reflect the current year. This commit updates the listed year to 2014.

see: http://www.copyright.gov/circs/circ01.pdf for more info
2014-03-12 12:06:58 -07:00
Igor Canadi
45ad75db80 Correct version of D16821 2014-03-12 09:38:59 -07:00
Igor Canadi
2b95dc1542 Revert "Fix bad merge of D16791 and D16767"
This reverts commit 839c8ecfcd.
2014-03-12 09:37:43 -07:00
Igor Canadi
5ba028c179 DBStress cleanup
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
2014-03-12 09:31:06 -07:00
sdong
839c8ecfcd Fix bad merge of D16791 and D16767
Summary: A bad Auto-Merge caused log buffer is flushed twice. Remove the unintended one.

Test Plan: Should already be tested (the code looks the same as when I ran unit tests).

Reviewers: haobo, igor

Reviewed By: haobo

CC: ljin, yhchiang, leveldb

Differential Revision: https://reviews.facebook.net/D16821
2014-03-11 21:31:57 -07:00
Lei Jin
86ba3e24e3 make assert based on FLAGS_prefix_size
Summary: as title

Test Plan: running python tools/db_crashtest.py

Reviewers: igor

Reviewed By: igor

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16803
2014-03-11 16:33:42 -07:00
sdong
bd45633b71 Fix data race against logging data structure because of LogBuffer
Summary:
@igor pointed out that there is a potential data race because of the way we use the newly introduced LogBuffer. After "bg_compaction_scheduled_--" or "bg_flush_scheduled_--", they can both become 0. As soon as the lock is released after that, DBImpl's deconstructor can go ahead and deconstruct all the states inside DB, including the info_log object hold in a shared pointer of the options object it keeps. At that point it is not safe anymore to continue using the info logger to write the delayed logs.

With the patch, lock is released temporarily for log buffer to be flushed before "bg_compaction_scheduled_--" or "bg_flush_scheduled_--". In order to make sure we don't miss any pending flush or compaction, a new flag bg_schedule_needed_ is added, which is set to be true if there is a pending flush or compaction but not scheduled because of the max thread limit. If the flag is set to be true, the scheduling function will be called before compaction or flush thread finishes.

Thanks @igor for this finding!

Test Plan: make all check

Reviewers: haobo, igor

Reviewed By: haobo

CC: dhruba, ljin, yhchiang, igor, leveldb

Differential Revision: https://reviews.facebook.net/D16767
2014-03-11 16:09:53 -07:00
Lei Jin
02dab3be11 fix db_stress test
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
2014-03-11 13:44:33 -07:00
sdong
6c66bc08d9 Temp Fix of LogBuffer flushing
Summary: To temp fix the log buffer flushing. Flush the buffer inside the lock. Clean the trunk before we find an eventual fix.

Test Plan: make all check

Reviewers: haobo, igor

Reviewed By: igor

CC: ljin, leveldb, yhchiang

Differential Revision: https://reviews.facebook.net/D16791
2014-03-11 11:37:40 -07:00
Igor Canadi
cb9802168f Add a comment after SignalAll()
Summary: Having code after SignalAll has already caused 2 bugs. Let's make sure this doesn't happen again.

Test Plan: no test

Reviewers: sdong, dhruba, haobo

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16785
2014-03-11 11:27:19 -07:00
sdong
01dcef114b Env to add a function to allow users to query waiting queue length
Summary: Add a function to Env so that users can query the waiting queue length of each thread pool

Test Plan: add a test in env_test

Reviewers: haobo

Reviewed By: haobo

CC: dhruba, igor, yhchiang, ljin, nkg-, leveldb

Differential Revision: https://reviews.facebook.net/D16755
2014-03-11 10:19:02 -07:00
Igor Canadi
d5de22dc09 Call PurgeObsoleteFiles() only when HaveSomethingToDelete()
Summary: as title

Test Plan: fixed the build failure http://ci-builds.fb.com/job/rocksdb_build/987/console

Reviewers: haobo

Reviewed By: haobo

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16743
2014-03-10 15:42:14 -07:00
Igor Canadi
9db8c4c556 Fix share_table_files bug
Summary: constructor wasn't properly constructing BackupableDBOptions

Test Plan: no test

Reviewers: benj

Reviewed By: benj

CC: leveldb

Differential Revision: https://reviews.facebook.net/D16749
2014-03-10 14:42:03 -07:00
sdong
fac58c0504 DBTest: remove perf_context's time > 0 check
Summary: DBTest checks  perf_context.seek_internal_seek_time > 0 and perf_context.find_next_user_entry_time > 0, which is not reliable. Remove them.

Test Plan: ./db_test

Reviewers: igor, haobo, ljin

Reviewed By: igor

CC: dhruba, yhchiang, leveldb

Differential Revision: https://reviews.facebook.net/D16737
2014-03-10 14:24:56 -07:00