846 Commits (a95323c021d026021678371c89bf0ac3ca38af84)

 

All Branches   

Search

Author	SHA1	Message	Date
Fabian Reinartz	76990518d3	Merge pull request #340 from prometheus/wal_migrate ... Migrate write ahead log	6 years ago
Fabian Reinartz	ee7ee059ef	Fix doc comments ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	a9a8fabd2c	Merge pull request #339 from prometheus/inittime ... Properly initialize head time	6 years ago
Fabian Reinartz	f8ec0074e7	Add Replace function ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	b81e0fbf2a	Address comments ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	45071c657c	Properly initialize head time ... This fixes various issues when initializing the head time range under different starting conditions. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	22fd3ef24e	Deal with zero-length segments ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	92e1b20957	Fix close handling ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	1a5573b4ce	Migrate write ahead log ... On startup, rewrite the old write ahead log into the new format once. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	3e76f0163e	Address comments ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	0ad2b8a349	docs: add new WAL format ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	3f538817f8	move WAL lock ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	d951140ab8	wal: avoid heap allocation in WAL reader ... The buffers we allocated were escaping to the heap, resulting in large memory usage spikes during startup and checkpointing in Prometheus. This attaches the buffer to the reader object to prevent this. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	7841d417b3	Ensure blocks are time-ordered in memory ... We assume in multiple places that the block list held by DB has blocks sequential by time. A regression caused us to hold them ordered by ULID, i.e. by creation time instead. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	def912ce0e	Integrate new WAL and checkpoints ... Remove the old WAL and drop in the new one Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	008399a6e0	Add checkpointing of WAL segments ... Create checkpoints from a sequence of WAL segments while filtering out obsolete data. The checkpoint format is again a sequence of WAL segments, which allows us to reuse the serialization format and implementation. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	449a2d0db7	wal: add segment type and repair procedure ... Allow to repair the WAL based on the error returned by a reader during a full scan over all records. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	8e1f97fad4	wal: add write ahead log package ... This adds a new WAL that's agnostic to the actual record contents. It's much simpler and should be more resilient than the existing one. Signed-off-by: Fabian Reinartz <freinartz@google.com>	6 years ago
Fabian Reinartz	99a2c4314f	Merge pull request #330 from codwu/tsdb-delete ... add rwmutex to prevent concurrent map read when delete series	6 years ago
codwu	bc6ef0b94e	rename `mts` to `intvlGroups` ... Signed-off-by: codwu <wuhan9087@163.com>	6 years ago
codwu	e4444ca48c	update `addInterval` function and test. ... Signed-off-by: codwu <wuhan9087@163.com>	6 years ago
codwu	667e539a7a	Merge branch 'master' of https://github.com/prometheus/tsdb into tsdb-delete	6 years ago
Fabian Reinartz	77db94c07e	Merge pull request #348 from BenoitKnecht/fix-block-boundaries ... Make sure blocks don't overlap to avoid outsider chunks	6 years ago
Benoît Knecht	24b223c161	db: add test for Querier returning too many blocks ... Due to the way blocks used to overlap by 1 millisecond (see #347), when requesting a 2-hour interval starting at `blocks[1].MinTime`, the `Querier` would consider three blocks: `blocks[0]`, `blocks[1]` and `blocks[2]`, because `blocks[0].MaxTime` and `blocks[2].MinTime` were in that interval. However, if the blocks don't overlap, only two blocks should be returned: `blocks[1]` and `blocks[2]`. This test ensures that it's indeed the case. Signed-off-by: Benoît Knecht <benoit.knecht@fsfe.org>	7 years ago
Benoît Knecht	1e1b2e163d	Make interval overlap comparisons more explicit ... Blocks are half-open intervals [a, b), while all other intervals (chunks, head, ...) are closed intervals [a, b]. Make that distinction explicit by defining `OverlapsClosedInterval()` methods for blocks and chunks, and using them in place of the more generic `intervalOverlap()` function. This change also fixes `db.Querier()` and `db.Delete()`, which could previously return one extraneous block at the end of the specified interval. Signed-off-by: Benoît Knecht <benoit.knecht@fsfe.org>	7 years ago
Benoît Knecht	4ed6b9ed72	db: add test for chunks that span beyond a block's boundaries ... Signed-off-by: Benoît Knecht <benoit.knecht@fsfe.org>	7 years ago
Benoît Knecht	0e4be5226a	db: block MaxTime should not be part of the block ... Block intervals are bound by `block.MinTime`, `block.MaxTime`, but they define a half-open interval: `[block.MinTime, block.MaxTime). However, when deciding if a chunk was part of a block or not, the `intervalOverlap()` function would consider both the chunk and the block intervals as being closed. Rather than modify the login in `intervalOverlap()`, we explicitly remove the last value from the interval when reading from head to persist blocks. Signed-off-by: Benoît Knecht <benoit.knecht@fsfe.org>	7 years ago
Fabian Reinartz	f87d00d78d	Merge pull request #356 from prometheus/logrollback ... Log series on rollback	7 years ago
Fabian Reinartz	ea607b9fc3	Log series on rollback ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	7 years ago
Fabian Reinartz	16727277e4	Merge pull request #354 from prometheus/cleandelete ... Add resilience to crashes during deletion	7 years ago
Fabian Reinartz	087c4c6d3a	Update doc comment ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	7 years ago
Fabian Reinartz	d907928e57	Clarify docs, error on unexpected meta read errors ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	7 years ago
Fabian Reinartz	af9003dcef	Add resilience to crashes during deletion ... Signed-off-by: Fabian Reinartz <freinartz@google.com>	7 years ago
Fabian Reinartz	0b200798fe	Merge pull request #344 from simonpasquier/fix-out-of-range-index ... chunks: fix potential "index out of range" error	7 years ago
Fabian Reinartz	0778d80ccf	Merge pull request #353 from cpatulea/patch-2 ... Update Appender.Add comment for uint64 refs.	7 years ago
Catalin Patulea	40766622ee	Update Appender.Add comment for uint64 refs. ... Follow-up to prometheus 0efecea6d4. Signed-off-by: Catalin Patulea <catalinp@google.com>	7 years ago
codwu	cd145c90d5	remove `put` function and use RLock in `Iter` function ... Signed-off-by: codwu <wuhan9087@163.com>	7 years ago
Simon Pasquier	f55ccd4ecb	Add unit tests ... Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago
Simon Pasquier	22061306aa	Address Julius's comment ... Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago
Simon Pasquier	e8fc6c8774	index: fix another Uvarint() return check ... Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago
codwu	84a45cb79a	add rwmutex to prevent concurrent map read when delete series ... Signed-off-by: codwu <wuhan9087@163.com>	7 years ago
Simon Pasquier	ee5fe8ea9f	chunks: fix potential "index out of range" error ... When binary.Uvarint() fails, the returned number of bytes is less than or equal to zero. Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago
Fabian Reinartz	c848349f07	Merge pull request #341 from codesome/delete-new-blocks-on-error ... Cleanup new blocks on 'CleanTombstones' faliure	7 years ago
Fabian Reinartz	9a96bc46de	Merge pull request #342 from prometheus/superq/promcon ... Add link to PromCon 2017 video.	7 years ago
Ganesh Vernekar	0c93850cd5	Merge pull request #1 from krasi-georgiev/pr/341 ... Add Test for Tombstone Cleaning after a failure	7 years ago
Krasi Georgiev	6094f35aa2	simplify if-else,test before the tombstone failure, more comments ... Signed-off-by: Krasi Georgiev <kgeorgie@redhat.com>	7 years ago
Krasi Georgiev	f31a0d6457	add Test for Tombstone Cleaning after a failure ... Signed-off-by: Krasi Georgiev <kgeorgie@redhat.com>	7 years ago
Ben Kochie	d5250f9673	Add link to PromCon 2017 video. ... Signed-off-by: Ben Kochie <superq@gmail.com>	7 years ago
Ganesh Vernekar	528439aa93	Cleanup new blocks on 'CleanTombstones' faliure. ... Signed-off-by: Ganesh Vernekar <cs15btech11018@iith.ac.in>	7 years ago
Simon Pasquier	eab78875fe	repair: improve error message ... This change adds the directory path to the error message in case the repair process fails. Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago