ecf1f99ddb
refactor: more flexible writer buffer config
...
This allows:
- different types (instead of guessing through the connection URL)
- sequencer counts (not used yet but will be by #2455 )
- extensible configs (e.g. to configure Kafka in a more granular way,
not wired up yet)
- future extensions (since we use a message now instead of a single
string)
**BREAKING: This requires changes for deployed systems / existing DBs!**
2021-09-02 16:41:35 +02:00
f81885c172
fix: limit kafka consumer queue to 10MB
...
I think that while this value limits the maximum memory consumption, it
is too optimistic.
2021-08-20 09:55:49 +02:00
6c0cefded3
fix: limit kafka consumer queue to 100MB
...
By default the queue will be filled to up to 1GB which makes finding
memory leaks quite hard.
2021-08-19 11:36:15 +02:00
817ed4b46b
feat: enable rdkafka statistics ( #2312 ) ( #2314 )
...
Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>
2021-08-17 09:28:02 +00:00
4a3fe01743
test: don't overdo it
2021-08-16 18:31:45 +02:00
5caa2ad8ec
fix: typo
2021-08-16 18:31:45 +02:00
1a7293015b
test: allow write buffer mocks that always fail
2021-08-16 18:27:09 +02:00
e3c263a006
test: add tooling to purge kafka topic data
2021-08-16 13:47:07 +02:00
21ebdee5a1
feat: make kafka topic creation code reusable
2021-08-16 13:47:07 +02:00
a72bacae67
test: use proper ignore instead of commenting out
2021-08-12 11:38:02 +02:00
a5c74f2798
feat: ability to inject mocked write buffers into server/database
2021-08-12 10:46:16 +02:00
55f68f0401
feat: add `type_name` to write buffer interface
2021-08-12 10:46:16 +02:00
32a5c8ca3a
feat: add ability to clear messages from mocked write buffers
2021-08-12 10:46:16 +02:00
3de6b44e23
build: use new rustdoc lint name ( #2261 )
...
* fix: nocache feature code rot
The MBChunk::snapshot code when using the "nocache" option no longer
compiles - this commit updates it to match the not(nocache) code.
* build: use updated broken_intra_doc_links name
The broken_intra_doc_links lint was renamed
rustdoc::broken_intra_doc_links
https://doc.rust-lang.org/rustdoc/lints.html
2021-08-11 19:48:51 +00:00
99b358d481
feat(iox): Enable snappy compression in the producer
2021-07-29 16:37:11 +02:00
eb310ecada
fix: ensure that the Kafka producer TS we report is also truncated
2021-07-28 15:48:02 +02:00
0fcec6b742
refactor: move ingest timestamp from sequence to sequended entry
2021-07-28 15:40:35 +02:00
d940d4f6d3
docs: explain how test timestamps are measured
2021-07-28 14:46:33 +02:00
e736bc6953
feat: add ingest timestamp to `Sequence`
...
This allows us to track wall-clock ingest time for entries that we
receive via write buffer (e.g. Kafka).
2021-07-28 14:41:18 +02:00
55490c279a
fix: Kafka watermark error for new partitions
2021-07-21 15:21:52 +02:00
5df88c70aa
feat: add ability to fetch watermarks from write buffer
2021-07-21 11:59:52 +02:00
58dd7e9532
Merge branch 'main' into crepererum/writer_buffer_seek
2021-07-20 12:29:18 +00:00
c01cfbc34c
fix: Increase kafka message size
2021-07-20 14:17:37 +02:00
ec7ebdff29
refactor: use lifetimes to ensure single stream / no seek while streaming
2021-07-20 13:52:33 +02:00
b0663a0337
feat: disallow multiple write buffer streams and seeking while streams
...
Multiple streams will mess up ordering. Seeking while streaming is
likely a bug and should not work.
2021-07-20 12:35:20 +02:00
38f4eec20e
feat: implement `seek` for write buffer
...
This is required to control replay ranges.
2021-07-20 10:25:56 +02:00
592424c896
refactor: use one stream per sequencer/partition
...
Advantages are:
- for large DBs w/ many partitions we can ingest data in-parallel
- on top of this change we can implement per-sequencer seeking, which is
required for replay
2021-07-19 12:26:58 +02:00
1676c33113
refactor: update write_buffer/src/kafka.rs
2021-07-16 16:56:03 +01:00
de0dce79f6
docs: Update write_buffer/src/kafka.rs
...
Co-authored-by: Edd Robinson <me@edd.io>
2021-07-16 17:53:02 +02:00
92623b5f32
fix: Set acks=all for kafka writes
2021-07-16 17:49:04 +02:00
cbadd65cfe
fix: Update write_buffer/src/kafka.rs
...
Co-authored-by: Andrew Lamb <alamb@influxdata.com>
2021-07-15 23:00:45 +02:00
06399e88e0
chore: Add some debug logs to write buffer
2021-07-15 22:18:03 +02:00
7301268b4f
fix: Increase the internal librdkafka producer queue size
...
Given that we've increased the max message size by a factor of 10, also
increase the internal producer queue max size by a factor of 10 to
reduce the number of retries needed to successfully enqueue messages to
Kafka.
Connects to #2007 .
2021-07-15 11:35:55 -04:00
fa3a2db0d3
fix: Retry adding Kafka messages to queue forever
...
By using [producer.send][] rather than [producer.send_result][] and
specifying Timeout::never.
Connects to #2007 .
[producer.send]: https://docs.rs/rdkafka/0.26.0/rdkafka/producer/future_producer/struct.FutureProducer.html#method.send
[producer.send_result]: https://docs.rs/rdkafka/0.26.0/rdkafka/producer/future_producer/struct.FutureProducer.html#method.send_result
2021-07-15 11:34:23 -04:00
a064820a70
fix: code comment should match the code
...
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
2021-07-15 17:20:46 +02:00
b5428e53a5
refactor: write buffer testing + better mocking
...
This refactors the write buffer a bit for:
- **Testing:** Add generic tests for the Kafka and the mocking
implementation. The same interface can be used easily add new
implementations (e.g. via Redis, filesystem, ...).
- **Partition on Write:** The caller of the writer operation must now
specify the partition/sequencer ID. The implicit partitioning of the
Kafka writer would have lead to broken data since we must never spill
entries w/ the same primary key over multiple partitions. At the
moment we will only use partition 0 but we can easily implement
better logic in the future.
- **Improved Mocking:** The mocked implementation now simulates a system
that feels more real. Especially the handling around multiple streams
and "write while read" has been improved. This will be helpful for
testing and for new features like seeking (during replay). A solid
realistic mock also helps us to ensure that the tests using the mock
do not rely on unrealistic behavior too much.
2021-07-15 17:20:45 +02:00
8d23dd6d6d
fix: Set kafka max message size in client
2021-07-14 14:46:49 +02:00
9cb9ae0874
chore: move write buffer into its own crate
2021-07-14 14:09:18 +02:00
f4a9a5ae56
fix: Remove write buffer
2021-06-04 14:40:17 -04:00
eddc9319ff
docs: deny broken intradoc links
2021-04-27 13:22:28 +02:00
938acb98e8
refactor: Rename WAL within the types, variables, and comments in the write_buffer crate
2021-04-21 17:43:03 +00:00
f0b93c5c8c
refactor: Rename the wal crate to write_buffer
2021-04-21 17:43:03 +00:00
48c43b136c
refactor: rename write_buffer --> mutable_buffer ( #595 )
...
* refactor: git mv write_buffer mutable_buffer
* refactor: update crate name references
* refactor: update some more references
2020-12-22 10:49:53 -05:00
263af1eeac
feat: implement read_group in the write_buffer ( #583 )
...
* feat: implement read_group in the write_buffer
* fix: Apply suggestions from code review
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
* fix: rustfmt
* fix: adjust tests for min/max
* fix: Update write_buffer/src/table.rs
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
2020-12-22 09:03:20 -05:00
28eac06d8f
refactor: Organize window_bounds the same as selector functions ( #594 )
...
* refactor: Organize window_bounds the same as selector functions
* fix: add missing file
2020-12-21 12:51:36 -05:00
bb96142564
chore: Update arrow dependencies, remove custom min/max implementation ( #585 )
...
* chore: Update arrow dependency
* fix: Update code for changes in datafusion
* fix: use arrow version of min_boolean
2020-12-21 12:31:39 -05:00
699504a022
feat: implement read_group for Sum, Count and Mean aggregates ( #557 )
...
* feat: Implement read_group for "normal" aggregates
* test: add tests
2020-12-15 09:35:00 -05:00
a6d2c13888
chore: Update arrow + other depenencies ( #540 )
...
* chore: Update arrow + other depenencies
* chore: Update write_buffer and query crate
2020-12-15 08:46:27 -05:00
6f473984d0
style: wrap comments
...
Runs rustfmt with the new config.
2020-12-11 18:22:26 +00:00
ea6b2f6bc8
refactor: remove minor code duplication ( #555 )
2020-12-11 11:18:00 -05:00
Marco Neumann
Raphael Taylor-Davies
Dom
Marko Mikulicic
kodiakhq[bot]
Edd Robinson
Carol (Nichols || Goulding)
Andrew Lamb