MSQ engine returns correct error codes for invalid user inputs in the query context. Also, using DruidExceptions for MSQ related errors happening in the Broker with improved error messages.
* Add aggregatorMergeStrategy property to SegmentMetadaQuery.
- Adds a new property aggregatorMergeStrategy to segmentMetadata query.
aggregatorMergeStrategy currently supports three types of merge strategies -
the legacy strict and lenient strategies, and the new latest strategy.
- The latest strategy considers the latest aggregator from the latest segment
by time order when there's a conflict when merging aggregators from different
segments.
- Deprecate lenientAggregatorMerge property; The API validates that both the new
and old properties are not set, and returns an exception.
- When merging segments as part of segmentMetadata query, the segments have a more
elaborate id -- <datasource>_<interval>_merged_<partition_number> format, similar to
the name format that segments usually contain. Previously it was simply "merged".
- Adjust unit tests to test the latest strategy, to assert the returned complete
SegmentAnalysis object instead of just the aggregators for completeness.
* Don't explicitly set strict strategy in tests
* Apply suggestions from code review
Co-authored-by: Katya Macedo <38017980+ektravel@users.noreply.github.com>
* Update docs/querying/segmentmetadataquery.md
* Apply suggestions from code review
Co-authored-by: Katya Macedo <38017980+ektravel@users.noreply.github.com>
---------
Co-authored-by: Katya Macedo <38017980+ektravel@users.noreply.github.com>
Uses a custom continusou jfr profiler.
Modifies the github actions for tests to do profiling only in the case
of jdk17, as the profiler requires jdk17+ to use the JFR streaming API
plus a few other language features in the code.
Continuous Profiling service is provided to the Apache Druid project
free of charge by Imply and any committer can request free access to
the UI.
* Fix a resource leak with Window processing
Additionally, in order to find the leak, there were
adjustments to the StupidPool to track leaks a bit better.
It would appear that the pool objects get GC'd during testing
for some reason which was causing some incorrect identification
of leaks from objects that had been returned but were GC'd along
with the pool.
* Suppress unused warning
* Add ZooKeeper connection state alerts and metrics.
- New metric "zk/connected" is an indicator showing 1 when connected,
0 when disconnected.
- New metric "zk/disconnected/time" measures time spent disconnected.
- New alert when Curator connection state enters LOST or SUSPENDED.
* Use right GuardedBy.
* Test fixes, coverage.
* Adjustment.
* Fix tests.
* Fix ITs.
* Improved injection.
* Adjust metric name, add tests.
Two changes:
1) Intern DecompressingByteBufferObjectStrategy. Saves ~32 bytes per column.
2) Split GenericIndexed into GenericIndexed.V1 and GenericIndexed.V2. The
major benefit here is isolating out the ByteBuffers that are only needed
for V2. This saves ~80 bytes for V1 (one buffer instead of two).
There are two ways of estimating heap footprint of an Aggregator:
1) AggregatorFactory#guessAggregatorHeapFootprint
2) AggregatorFactory#factorizeWithSize + Aggregator#aggregateWithSize
When the second path is used, the default implementation of factorizeWithSize
is now updated to delegate to guessAggregatorHeapFootprint, making these equivalent.
The old logic used getMaxIntermediateSize, which is less accurate.
Also fixes a bug where, when using the second path, calling factorizeWithSize
on PassthroughAggregatorFactory would fail because getMaxIntermediateSize was
not implemented. (There is no buffer aggregator, so there would be no need.)
Cache is disabled for GroupByStrategyV2 on broker since the pr #3820 [groupBy v2: Results not fully merged when caching is enabled on the broker]. But we can enable the result-level cache on broker for GroupByStrategyV2 and keep the segment-level cache disabled.
* Claim full support for Java 17.
No production code has changed, except the startup scripts.
Changes:
1) Allow Java 17 without DRUID_SKIP_JAVA_CHECK.
2) Include the full list of opens and exports on both Java 11 and 17.
3) Document that Java 17 is both supported and preferred.
4) Switch some tests from Java 11 to 17 to get better coverage on the
preferred version.
* Doc update.
* Update errorprone.
* Update docker_build_containers.sh.
* Update errorprone in licenses.yaml.
* Add some more run-javas.
* Additional run-javas.
* Update errorprone.
* Suppress new errorprone error.
* Add exports and opens in ForkingTaskRunner for Java 11+.
Test, doc changes.
* Additional errorprone updates.
* Update for errorprone.
* Restore old fomatting in LdapCredentialsValidator.
* Copy bin/ too.
* Fix Java 15, 17 build line in docker_build_containers.sh.
* Update busybox image.
* One more java command.
* Fix interpolation.
* IT commandline refinements.
* Switch to busybox 1.34.1-glibc.
* POM adjustments, build and test one IT on 17.
* Additional debugging.
* Fix silly thing.
* Adjust command line.
* Add exports and opens one more place.
* Additional harmonization of strong encapsulation parameters.
One of the most requested features in druid is to have an ability to download big result sets.
As part of #14416 , we added an ability for MSQ to be queried via a query friendly endpoint. This PR builds upon that work and adds the ability for MSQ to write select results to durable storage.
We write the results to the durable storage location <prefix>/results/<queryId> in the druid frame format. This is exposed to users by
/v2/sql/statements/:queryId/results.
* Fix ColumnSignature error message and jdk17 test issue.
On jdk17, the "problem" part of the error message could change from
NullPointerException to:
Cannot invoke "String.length()" because "s" is null
Due to the new more-helpful NPEs in Java 17. This broke the expectation
and led to test failures on this case.
This patch fixes the problem by improving the error message so it isn't
a generic NullPointerException.
* Fix format.
This commit borrows some test definitions from Drill's test suite
and tries to use them to flesh out the full validation of window
function capbilities.
In order to be able to run these tests, we also add the ability to
run a Scan operation against segments, which also meant an
implementation of RowsAndColumns for frames.
UniformGranularityTest's test to test a large number of intervals
runs through 10 years of 1 second intervals. This pushes a lot of
stuff through IntervalIterator and shows up in terms of test
runtime as one of the hottest tests. Most of the time is going to
constructing jodatime objects because it is doing things with
DateTime objects instead of millis. Change the calls to use
millis instead and things go faster.
If a server is removed during `HttpServerInventoryView.serverInventoryInitialized`,
the initialization gets stuck as this server is never synced. The method eventually times
out (default 250s).
Fix: Mark a server as stopped if it is removed. `serverInventoryInitialized` only waits for
non-stopped servers to sync.
Other changes:
- Add new metrics for better debugging of slow broker/coordinator startup
- `segment/serverview/sync/healthy`: whether the server view is syncing properly with a server
- `segment/serverview/sync/unstableTime`: time for which sync with a server has been unstable
- Clean up logging in `HttpServerInventoryView` and `ChangeRequestHttpSyncer`
- Minor refactor for readability
- Add utility class `Stopwatch`
- Add tests and stubs
* combine string column implementations
changes:
* generic indexed, front-coded, and auto string columns now all share the same column and index supplier implementations
* remove CachingIndexed implementation, which I think is largely no longer needed by the switch of many things to directly using ByteBuffer, avoiding the cost of creating Strings
* remove ColumnConfig.columnCacheSizeBytes since CachingIndexed was the only user
* Add "stringEncoding" parameter to DataSketches HLL.
Builds on the concept from #11172 and adds a way to feed HLL sketches
with UTF-8 bytes.
This must be an option rather than always-on, because prior to this
patch, HLL sketches used UTF-16LE encoding when hashing strings. To
remain compatible with sketch images created prior to this patch -- which
matters during rolling updates and when reading sketches that have been
written to segments -- we must keep UTF-16LE as the default.
Not currently documented, because I'm not yet sure how best to expose
this functionality to users. I think the first place would be in the SQL
layer: we could have it automatically select UTF-8 or UTF-16LE when
building sketches at query time. We need to be careful about this, though,
because UTF-8 isn't always faster. Sometimes, like for the results of
expressions, UTF-16LE is faster. I expect we will sort this out in
future patches.
* Fix benchmark.
* Fix style issues, improve test coverage.
* Put round back, to make IT updates easier.
* Fix test.
* Fix issue with filtered aggregators and add test.
* Use DS native update(ByteBuffer) method. Improve test coverage.
* Add another suppression.
* Fix ITAutoCompactionTest.
* Update benchmarks.
* Updates.
* Fix conflict.
* Adjustments.
In these other cases, stick to plain "filter". This simplifies lots of
logic downstream, and doesn't hurt since we don't have intervals-specific
optimizations outside of tables.
Fixes an issue where we couldn't properly filter on a column from an
external datasource if it was named __time.
* Properly read SQL-compatible segments in default-value mode.
Main changes:
1) Dictionary-encoded and front-coded string columns: in default-value
mode, detect cases where a dictionary has the empty string in it, then
either combine it with null (if null is present) or replace it with
null (if null is not present).
2) Numeric nullable columns: in default-value mode, ignore the null
value bitmap. This causes all null numbers to be read as zeroes.
Testing strategy:
1) Add a mmappedWithSqlCompatibleNulls case to BaseFilterTest that
writes segments under SQL-compatible mode, and reads them under
default-value mode.
2) Unit tests for the new wrapper classes (CombineFirstTwoEntriesIndexed,
CombineFirstTwoValuesColumnarInts, CombineFirstTwoValuesColumnarMultiInts,
CombineFirstTwoValuesIndexedInts).
* Fix a mistake, use more singlethreadedness.
* WIP
* Tests, improvements.
* Style.
* See Spot bug.
* Remove unused method.
* Address review comments.
1) Read bitmaps even if we don't retain them.
2) Combine StringFrontCodedDictionaryEncodedColumn and ScalarStringDictionaryEncodedColumn.
* Add missing tests.
This PR aims to expose a new API called
"@path("/druid/v2/sql/statements/")" which takes the same payload as the current "/druid/v2/sql" endpoint and allows users to fetch results in an async manner.
* Fix another infinite loop and remove Mockito usage
The ConfigManager objects were `started()` without ever being
stopped. This scheduled a poll call that never-ended, to make
matters worse, the poll interval was set to 0 ms, making an
infinite poll with 0 sleep, i.e. an infinite loop.
Also introduce test classes and remove usage of mocks
* Checkstyle
Adds support for automatic cleaning of a "query-results" directory in durable storage. This directory will be cleaned up only if the task id is not known to the overlord. This will allow the storage of query results after the task has finished running.
* Cache parsed expressions and binding analysis in more places.
Main changes:
1) Cache parsed and analyzed expressions within PlannerContext for a
single SQL query.
2) Cache parsed expressions together with input binding analysis using
a new class AnalyzeExpr.
This speeds up SQL planning, because SQL planning involves parsing
analyzing the same expression strings over and over again.
* Fixes.
* Fix style.
* Fix test.
* Simplify: get rid of AnalyzedExpr, focus on caching.
* Rename parse -> parseExpression.
Changes:
- Throw an `InsertCannotAllocateSegmentFault` if the allocated segment is not aligned with
the requested granularity.
- Tests to verify new behaviour
Users can now add a guardrail to prevent subquery’s results from exceeding the set number of bytes by setting druid.server.http.maxSubqueryRows in Broker's config or maxSubqueryRows in the query context. This feature is experimental for now and would default back to row-based limiting in case it fails to get the accurate size of the results consumed by the query.
Recently, we have seen flakiness in these two tests, apparently due to
computations based on Runtime.getRuntime().maxMemory() differing during
static initialization and in the actual tests. I can't think of a reason
why this would be happening, but anyway, this patch switches the tests to
use the statics instead of recomputing Runtime.getRuntime().maxMemory().
* Fix compatibility issue with SqlTaskResource
The DruidException changes broke the response format
for errors coming back from the SqlTaskResource, so fix those
Added a new monitor SysMonitorOshi to replace SysMonitor. The new monitor has a wider support for different machine architectures including ARM instances. Please switch to SysMonitorOshi as SysMonitor is now deprecated and will be removed in future releases.
This commit does a complete revamp of the coordinator to address problem areas:
- Stability: Fix several bugs, add capabilities to prioritize and cancel load queue items
- Visibility: Add new metrics, improve logs, revamp `CoordinatorRunStats`
- Configuration: Add dynamic config `smartSegmentLoading` to automatically set
optimal values for all segment loading configs such as `maxSegmentsToMove`,
`replicationThrottleLimit` and `maxSegmentsInNodeLoadingQueue`.
Changed classes:
- Add `StrategicSegmentAssigner` to make assignment decisions for load, replicate and move
- Add `SegmentAction` to distinguish between load, replicate, drop and move operations
- Add `SegmentReplicationStatus` to capture current state of replication of all used segments
- Add `SegmentLoadingConfig` to contain recomputed dynamic config values
- Simplify classes `LoadRule`, `BroadcastRule`
- Simplify the `BalancerStrategy` and `CostBalancerStrategy`
- Add several new methods to `ServerHolder` to track loaded and queued segments
- Refactor `DruidCoordinator`
Impact:
- Enable `smartSegmentLoading` by default. With this enabled, none of the following
dynamic configs need to be set: `maxSegmentsToMove`, `replicationThrottleLimit`,
`maxSegmentsInNodeLoadingQueue`, `useRoundRobinSegmentAssignment`,
`emitBalancingStats` and `replicantLifetime`.
- Coordinator reports richer metrics and produces cleaner and more informative logs
- Coordinator uses an unlimited load queue for all serves, and makes better assignment decisions
Introduce DruidException, an exception whose goal in life is to be delivered to a user.
DruidException itself has javadoc on it to describe how it should be used. This commit both introduces the Exception and adjusts some of the places that are generating exceptions to generate DruidException objects instead, as a way to show how the Exception should be used.
This work was a 3rd iteration on top of work that was started by Paul Rogers. I don't know if his name will survive the squash-and-merge, so I'm calling it out here and thanking him for starting on this.
Description:
Druid allows a configuration of load rules that may cause a used segment to not be loaded
on any historical. This status is not tracked in the sys.segments table on the broker, which
makes it difficult to determine if the unavailability of a segment is expected and if we should
not wait for it to be loaded on a server after ingestion has finished.
Changes:
- Track replication factor in `SegmentReplicantLookup` during evaluation of load rules
- Update API `/druid/coordinator/v1metadata/segments` to return replication factor
- Add column `replication_factor` to the sys.segments virtual table and populate it in
`MetadataSegmentView`
- If this column is 0, the segment is not assigned to any historical and will not be loaded.
* fix kafka input format reader schema discovery and partial schema discovery to actually work right, by re-using dimension filtering logic of MapInputRowParser
changes:
* auto columns no longer participate in generic 'null column' handling, this was a mistake to try to support and caused ingestion failures due to mismatched ColumnFormat, and will be replaced in the future with nested common format constant column functionality (not in this PR)
* fix bugs with auto columns which contain empty objects, empty arrays, or primitive types mixed with either of these empty constructs
* fix bug with bound filter when upper is null equivalent but is strict
Changes
- Add a `DruidException` which contains a user-facing error message, HTTP response code
- Make `EntryExistsException` extend `DruidException`
- If metadata store max_allowed_packet limit is violated while inserting a new task, throw
`DruidException` with response code 400 (bad request) to prevent retries
- Add `SQLMetadataConnector.isRootCausePacketTooBigException` with impl for MySQL
The class apparently only exists to add a toString()
method to Indexes, which basically just crashes any debugger
on any meaningfully sized index. It's a pointless
abstract class that basically only causes pain.
In this PR, we are enhancing KafkaEmitter, to emit metadata about published segments (SegmentMetadataEvent) into a Kafka topic. This segment metadata information that gets published into Kafka, can be used by any other downstream services to query Druid intelligently based on the segments published. The segment metadata gets published into kafka topic in json string format similar to other events.
### Description
This change allows for consideration of the input format and compression when computing how to split the input files among available tasks, in MSQ ingestion, when considering the value of the `maxInputBytesPerWorker` query context parameter. This query parameter allows users to control the maximum number of bytes, with granularity of input file / object, that ingestion tasks will be assigned to ingest. With this change, this context parameter now denotes the estimated weighted size in bytes of the input to split on, with consideration for input format and compression format, rather than the actual file size, reported by the file system. We assume uncompressed newline delimited json as a baseline, with scaling factor of `1`. This means that when computing the byte weight that a file has towards the input splitting, we take the file size as is, if uncompressed json, 1:1. It was found during testing that gzip compressed json, and parquet, has scale factors of `4` and `8` respectively, meaning that each byte of data is weighted 4x and 8x respectively, when computing input splits. This weighted byte scaling is only considered for MSQ ingestion that uses either LocalInputSource or CloudObjectInputSource at the moment. The default value of the `maxInputBytesPerWorker` query context parameter has been updated from 10 GiB, to 512 MiB
Laksh Singla
Abhishek Radhakrishnan
Sam Rash
imply-cheddar
Gian Merlino
hqx871
Kashif Faraz
Karan Kumar
Clint Wylie
Adarsh Sanjeev
Hardik Bajaj
George Shiqi Wu
Pranav
Harini Rajendran
zachjsh