Add query context parameter for segment load wait #15076
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
github-actions
bot
added
Area - Documentation
Area - Batch Ingestion
Area - MSQ
cryptoe
reviewed
Oct 5, 2023
docs/multi-stage-query/reference.md
Outdated
| @@ -246,6 +246,7 @@ The following table lists the context parameters for the MSQ task engine: | |||
| | `durableShuffleStorage` | SELECT, INSERT, REPLACE <br /><br />Whether to use durable storage for shuffle mesh. To use this feature, configure the durable storage at the server level using `druid.msq.intermediate.storage.enable=true`). If these properties are not configured, any query with the context variable `durableShuffleStorage=true` fails with a configuration error. <br /><br /> | `false` | | |||
| | `faultTolerance` | SELECT, INSERT, REPLACE<br /><br /> Whether to turn on fault tolerance mode or not. Failed workers are retried based on [Limits](#limits). Cannot be used when `durableShuffleStorage` is explicitly set to false. | `false` | | |||
| | `selectDestination` | SELECT<br /><br /> Controls where the final result of the select query is written. <br />Use `taskReport`(the default) to write select results to the task report. <b> This is not scalable since task reports size explodes for large results </b> <br/>Use `durableStorage` to write results to durable storage location. <b>For large results sets, its recommended to use `durableStorage` </b>. To configure durable storage see [`this`](#durable-storage) section. | `taskReport` | | |||
| | `segmentLoadWait` | INSERT, REPLACE<br /><br /> Whether the controller should wait for segments to be loaded before exiting. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query. | `false` | | |||
There was a problem hiding this comment.
segmentHandedOff ? How does this name sound ?
| @@ -159,29 +159,18 @@ public void waitForSegmentsToLoad() | |||
| return; | |||
| } | |||
|
|
|||
| Iterator<String> iterator = versionsToAwait.iterator(); | |||
| log.info( | |||
| log.debug( | |||
There was a problem hiding this comment.
Can we log.info this every minute ?
There was a problem hiding this comment.
And update the log message to say that even if the task is cancelled, the segments will continue to load.
| { | ||
| Request request = brokerClient.makeRequest(HttpMethod.POST, "/druid/v2/sql/"); | ||
| SqlQuery sqlQuery = new SqlQuery(StringUtils.format(LOAD_QUERY, datasource, version), | ||
| SqlQuery sqlQuery = new SqlQuery(StringUtils.format(LOAD_QUERY, datasource, versionsInClauseString), |
There was a problem hiding this comment.
While running ut's I also saw
2023-10-05T11:21:28,837 WARN [query-2cd60052-3011-47ae-87be-ec47100f649d-segment-load-waiter-0] org.apache.druid.msq.exec.SegmentLoadStatusFetcher - Exception occurred while waiting for segments to load. Exiting.
java.lang.NullPointerException: null
at org.apache.druid.msq.exec.SegmentLoadStatusFetcher.fetchLoadStatusForVersion(SegmentLoadStatusFetcher.java:262) ~[classes/:?]
at org.apache.druid.msq.exec.SegmentLoadStatusFetcher.lambda$waitForSegmentsToLoad$0(SegmentLoadStatusFetcher.java:174) ~[classes/:?]
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515) ~[?:?]
at com.google.common.util.concurrent.TrustedListenableFutureTask$TrustedFutureInterruptibleTask.runInterruptibly(TrustedListenableFutureTask.java:131) ~[guava-31.1-jre.jar:?]
at com.google.common.util.concurrent.InterruptibleTask.run(InterruptibleTask.java:74) ~[guava-31.1-jre.jar:?]
at com.google.common.util.concurrent.TrustedListenableFutureTask.run(TrustedListenableFutureTask.java:82) ~[guava-31.1-jre.jar:?]
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) ~[?:?]
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) ~[?:?]
at java.lang.Thread.run(Thread.java:829) ~[?:?]
2023-10-05T11:21:28,846 INFO [main] org.apache.druid.msq.exec.WorkerImpl - Stopping gracefully for taskId [query-2cd60052-3011-47ae-87be-ec47100f649d-worker0_0]
2023-10-05T11:21:28,848 INFO [main] org.apache.druid.msq.test.MSQTestBase - found generated segments: DataSegment{binaryVersion=9, id=foo1_-146136543-09-08T08:23:32.096Z_146140482-04-24T15:36:27.903Z_test, loadSpec={type=>local, path=>/var/folders/sx/n0v16tnj6mvf6hd14l6j6dth0000gn/T/junit8066366800031570368/localsegments/foo1/-146136543-09-08T08:23:32.096Z_146140482-04-24T15:36:27.903Z/test/0/index/}, dimensions=[dim_mv], metrics=[], shardSpec=NumberedShardSpec{partitionNum=0, partitions=100}, lastCompactionState=null, size=1037}
2023-10-05T11:21:28,859 INFO [main] org.apache.druid.msq.test.MSQTestBase - Found spec: {
"query" : {
Lets fix it in this PR as well.
LakshSingla
reviewed
Oct 5, 2023
docs/multi-stage-query/reference.md
Outdated
| @@ -246,6 +246,7 @@ The following table lists the context parameters for the MSQ task engine: | |||
| | `durableShuffleStorage` | SELECT, INSERT, REPLACE <br /><br />Whether to use durable storage for shuffle mesh. To use this feature, configure the durable storage at the server level using `druid.msq.intermediate.storage.enable=true`). If these properties are not configured, any query with the context variable `durableShuffleStorage=true` fails with a configuration error. <br /><br /> | `false` | | |||
| | `faultTolerance` | SELECT, INSERT, REPLACE<br /><br /> Whether to turn on fault tolerance mode or not. Failed workers are retried based on [Limits](#limits). Cannot be used when `durableShuffleStorage` is explicitly set to false. | `false` | | |||
| | `selectDestination` | SELECT<br /><br /> Controls where the final result of the select query is written. <br />Use `taskReport`(the default) to write select results to the task report. <b> This is not scalable since task reports size explodes for large results </b> <br/>Use `durableStorage` to write results to durable storage location. <b>For large results sets, its recommended to use `durableStorage` </b>. To configure durable storage see [`this`](#durable-storage) section. | `taskReport` | | |||
| | `segmentLoadWait` | INSERT, REPLACE<br /><br /> Whether the controller should wait for segments to be loaded before exiting. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query. | `false` | | |||
There was a problem hiding this comment.
seems very geared toward developers, and not the users. We can remove the fact that it queries the broker (since that is an implementation detail).
WDYT of something along the lines:
Suggested change
| | `segmentLoadWait` | INSERT, REPLACE<br /><br /> Whether the controller should wait for segments to be loaded before exiting. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query. | `false` | | |
| | `segmentLoadWait` | INSERT, REPLACE<br /><br /> If set, the ingest query waits for the generated segment to be loaded before exiting, else the ingest query exits without waiting. The task and live reports contain the information (about ??) if this flag is set. The drawback is that the tasks stall till the segments are loaded, however ensures that... (advantage) | `false` | |
abhishekagarwal87
pushed a commit
that referenced
this pull request
Oct 11, 2023
LakshSingla
pushed a commit
to LakshSingla/druid
that referenced
this pull request
Oct 14, 2023
This relies on the work done in apache#14322 and apache#15076. It allows the user to set waitTillSegmentsLoad in the query context (if they want, else it defaults to true) and shows the results in the UI :
LakshSingla
added a commit
that referenced
this pull request
Oct 16, 2023
This relies on the work done in #14322 and #15076. It allows the user to set waitTillSegmentsLoad in the query context (if they want, else it defaults to true) and shows the results in the UI : Co-authored-by: Sébastien <[email protected]>
ektravel
pushed a commit
to ektravel/druid
that referenced
this pull request
Oct 16, 2023
Add segmentLoadWait as a query context parameter. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query.
ektravel
pushed a commit
to ektravel/druid
that referenced
this pull request
Oct 16, 2023
This relies on the work done in apache#14322 and apache#15076. It allows the user to set waitTillSegmentsLoad in the query context (if they want, else it defaults to true) and shows the results in the UI :
CaseyPan
pushed a commit
to CaseyPan/druid
that referenced
this pull request
Nov 17, 2023
Add segmentLoadWait as a query context parameter. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query.
CaseyPan
pushed a commit
to CaseyPan/druid
that referenced
this pull request
Nov 17, 2023
This relies on the work done in apache#14322 and apache#15076. It allows the user to set waitTillSegmentsLoad in the query context (if they want, else it defaults to true) and shows the results in the UI :
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
#14322 added the feature of allowing the waiting for segments to be loaded to the MSQ controller. This adds a context parameter to control if this behaviour should be used. It also optimizes the query to make fewer calls to the broker by clubbing all the versions into the same request.
Notes:
waitTillSegmentsLoadas a query context parameter. If this is true, the controller queries the broker and waits till the segments created (if any) have been loaded by the load rules. The controller also provides this information in the live reports and task reports. If this is false, the controller exits immediately after finishing the query.This PR has: