276 Commits (main)

Author	SHA1	Message	Date
Ivan Kalita	08e3c4385f	feat(metastore): shard sections queries over index files (#20134) ... Running metastore queriesin a distributed manner using query engine workers: - Build a distributed metastore plan from GetIndexes and execute it via the v2 scheduler/worker pipeline - Split the request into per-index PointersScan tasks, then fan-in and CollectSections to produce final section descriptors - Add physical plan + protobuf support for Merge/PointersScan, improve tracing/cleanup, and add coverage for planner/workflow/proto roundtrips	4 days ago
Ivan Kalita	9be17c7a42	feat(metastore): metastore DI (#20253) ... This commit fixes a problem that happens when we instantiate 2 metastores in a single process. Even though this can't happen at the moment it will be the case once we introduce distributed metastore queries - when running both schedulers and workers in the same process all of them will need an instance of metastore. Each instance (before this PR) internally creates and registers metastore metrics which causes panic due to exactly the same metrics were already registered by the same registerer. I attempt to fix this problem via creating and registering metastore metrics only once.	1 week ago
Ivan Kalita	e4ec8441d3	feat(metastore): use arrow for scanning and blooms (#20234) ... Changing metastore implementation to work with arrow.RecordBatch without intermediate materialization to structs. metastore.Sections now uses new metastore.GetIndexes, metastore.IndexSectionsReader and metastore.CollectSections. We'd like to run metastore queries in a distributed manner using engine v2. In order to achieve that we need to split the work that metastore.Sections does internally, schedule it on different workers, and collect the results. The engine achieves that via pipelines that produce arrow.RecordBatch-es. In this PR the metastore.Sections is updated such a way to make it easier to integrate metastore with the engine.	1 week ago
Ashwanth	7da7f000ff	chore: use projection for indexobj streams reader (#20205)	2 weeks ago
George Robinson	a03e00a3bc	chore: better comments when end offset is zero (#20196)	2 weeks ago
George Robinson	333da739c7	feat: add processed records metric (#20191)	2 weeks ago
George Robinson	c0c27b3596	feat: add downscalePermittedFunc to check downscale is permitted (#20171)	2 weeks ago
Ashwanth	54f9723af1	chore(xcap): adds exporter to summarise capture as a structured log line (#20099)	3 weeks ago
Ashwanth	728b308edf	chore: update read stats to use xcap (#20095)	3 weeks ago
Ashwanth	cbad9d4ccb	chore(xcap): support events, status and improve coverage (#20096)	3 weeks ago
Ivan Kalita	827292d8ed	chore(metastore): add columnar indexpointers reader (#20068) ... Similar to https://github.com/grafana/loki/pull/20053 and https://github.com/grafana/loki/pull/19992 I add a columnar reader for index pointers sections.	3 weeks ago
Ivan Kalita	ee58c333b3	chore(metastore): use column reader for blooms (#20053) ... This PR is a continuation of https://github.com/grafana/loki/pull/19992. I use `pointers.Reader` to refine the sections that match bloom filters. I also fix minor issues that I introduced in the previous PR: - convert the columns only once per column (instead of doing it for every row) - ignore the sections that are of a wrong "type"	4 weeks ago
benclive	225a0595ac	chore: Define an E2E ingestion processing time metric (#20054)	4 weeks ago
Ivan Kalita	3559d4bd42	chore(metastore): add column reader for pointers section (#19992) ... Adding columnar reader for (log)pointers section. Using it to read the stream section pointers. New reader is slower than existing RowReader due to double-translation of the data (columns->rows->rows vs columns->rows->columns->rows). The idea though is to get rid of the double-translation in the future.	4 weeks ago
benclive	044ded2e29	chore: Split indexing into distinct steps (#19987)	4 weeks ago
benclive	2ce207fbdb	chore: Use shared zstd encoders during dataobj build (#20044)	4 weeks ago
Christian Haudum	8b59b944f2	chore(dataobj): Use common base config for indexobj and logsobj builders (#20037) ... Use the same config struct for both the indexobj builder and the logsobj builder, since they are configured identically, only with different default values.	4 weeks ago
Robert Fratto	806bcd5dd7	chore(dataobj): stop preallocating memory for page builders (#20043)	4 weeks ago
George Robinson	677b2ecda1	feat: add prepare downscale handler (#20007)	1 month ago
George Robinson	9fb90309c0	chore: clean up prepare downscale handler (#20006)	1 month ago
Periklis Tsirakidis	32dbef99b8	fix(dataobj): Flush into multiple index objects when ErrBuilderFull (#19223)	1 month ago
benclive	bda352b67d	chore: Use flag to set index consumer group (#19965)	1 month ago
Paul Rogers	91d4eb6ce1	feat: Add partition state to consumption lag metric (#19912)	1 month ago
benclive	17285207f1	chore: Reduce buffer size for bloom lookups (#19926)	1 month ago
benclive	9c552a23eb	chore: Add a dataobj readme (#19910)	1 month ago
Christian Haudum	1cc97513d7	chore(metastore): Cleanup and reduce log volume (#19889) ... Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	1 month ago
Christian Haudum	97fc8b3a50	chore(metastore): Simplify public API of Metastore (#19886) ... Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	1 month ago
renovate-sh-app[bot]	d76b3bf495	fix(deps): update module github.com/apache/arrow-go/v18 to v18.4.1 (main) (#19750) ... Signed-off-by: renovate-sh-app[bot] <219655108+renovate-sh-app[bot]@users.noreply.github.com> Co-authored-by: renovate-sh-app[bot] <219655108+renovate-sh-app[bot]@users.noreply.github.com> Co-authored-by: Paul Rogers <paul.rogers@grafana.com>	2 months ago
George Robinson	a28e755e79	chore: rename the prepare delayed downscale handler to better communicate what it does (#19727)	2 months ago
George Robinson	0af297cd4c	chore: remove loki_dataobj_consumer_processing_delay histogram (#19718)	2 months ago
Christian Haudum	0b4a7f5a2c	chore(dataobj-consumer): Re-use builder for CopyAndSort() (#19696) ... Re-use logsobj.Builder instance in partition processor when sorting logs object wide after flushing. The metrics for the builder in CopyAndSort are never registered, which means the loki_dataobj_sort_duration_seconds histogram is never exported/scraped. We cannot register them with RegisterMetrics() because the metrics are already registered here, and registering them twice would cause a panic.	2 months ago
benclive	6ee50064ea	chore: Ensure partition is set on partition-processor (#19693)	2 months ago
George Robinson	da37290a40	fix: loki_dataobj_sort_duration_seconds never registered (#19679)	2 months ago
Christian Haudum	73bc30d9a8	perf(engine): Improve regexp expression evaluation (#19644) ... This PR changes the evaluation of regex-match binary expression evaluations so that the regexp on right hand side of the expression is only compiled once for each batch evaluation, rather than for each row. This saves a ton of allocations and improves general performance of queries with a regexp line or label filter.	2 months ago
George Robinson	d146c9f4ad	feat: add enabled flag for dataobjs (#19613)	2 months ago
Ashwanth	5a06d44826	chore: table build fill all columns to max rows (#19614)	2 months ago
benclive	a6de577b3a	chore: Fix page skipping with empty predicates (#19602)	2 months ago
Ashwanth	bfd70602dd	chore: correct dataobj reader rows read return value (#19583)	2 months ago
George Robinson	977fb2ba6f	feat: consume single topics in dataobj-consumer using partition ring (#19186)	2 months ago
benclive	ea2b4d0f84	chore: Use Kafka round-robin balancer for index-builder (#19574)	2 months ago
Stas Spiridonov	160dc2c493	chore: removed arrow-go allocators/retain/release (#19569)	2 months ago
benclive	c31e25762e	chore: Fix metadata bloom check with multiple predicates (#19389)	2 months ago
Ashwanth	fa8bd848f6	chore: update buildTable to sort rows using `BuilderOptions.SortOrder` (#19515)	2 months ago
Robert Fratto	7556192423	chore(dataobj): use single Zstd decoder (#19508)	2 months ago
Christian Haudum	c06eb636e5	chore(engine): Add "compatibility node" to physical plan to adhere with naming of "colliding labels" in v1 engine (#19470) ... ### Summary The v1 engine has a mechanism to rename labels in case they have the same name but different origin, such as labels, structured metadata, or parsed fields. 1. In case a log line has a structured metadata key with the same name as the label name of the stream, than the metadata key is suffixed with `_extracted`, such as `service_extracted`, if `service` exists in both `labels` and `metadata`. 2. In case a parser creates a parsed field with the same as the label name of the stream, then the parsed key is suffixed with `_extracted` in the same way as case 1. However, if the field name also collides with a structured metadata key, then the extracted structured metadata is replaced with the extracted parsed field. This PR only implements the first case. As a follow up PR, the second case needs to be implemented as well. Additionally, the newly introduced "compatibility node" should also be made optional with a feature flag and/or per-request. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 months ago
benclive	b0f5ca7f8c	chore: Dedup log objects during build (#19378)	3 months ago
Christian Haudum	386d4e1653	chore(dataobj): Make logs sort order in dataobjects configurable (#19373) ... This PR introduces a new configuration option `-dataobj-consumer.dataobj-sort-order` which controls the sort order of logs in the logs section of the dataobj when building new objects. There are two available sort options: 1. `timestamp DESC, streamID ASC` (configuration value `timestamp-desc`) 2. `streamID ASC, timestamp DESC` (configuration value `stream-asc`) The new setting is undocumented and may be removed in the future without notice. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 months ago
benclive	c5a0a3827d	chore: Memory improvements in dataobj page downloads (#19301)	3 months ago
benclive	2dac5ef8a7	chore: Write empty message values in logs section writer (#19359)	3 months ago
Sophie Waldman	af6983e9de	fix: Add special case handling for comparing empty and nil values (#19348) ... We can have a predicate checking for an empty string, e.g. {"foo" = ""}. If a row has no value for the column "foo", it will return a nil value for that column. Thus, when we go to compare the row with the predicate, we will be comparing an empty value to a nil value. This change adjusts the comparison check so that we correctly handle this special case.	3 months ago