You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Caleb Rackliffe (Jira)" <ji...@apache.org> on 2025/03/05 22:47:00 UTC

[jira] [Commented] (CASSANDRA-19596) Improve IntervalTree build throughput

    [ https://issues.apache.org/jira/browse/CASSANDRA-19596?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17932780#comment-17932780 ] 

Caleb Rackliffe commented on CASSANDRA-19596:
---------------------------------------------

I just ran the Python dtests against 5.0 while investigating CI failures for CASSANDRA-20397...

{noformat}
pytest --cassandra-dir=/Users/maedhroz/Projects/Apache/asf-cassandra scrub_test.py::TestScrub
{noformat}

...and I'm seeing this:

{noformat}
test_standalone_scrub_essential_files_only failed and was not selected for rerun.
	<class 'AssertionError'>
	Attempted to check that stderr was empty. Instead, stderr is WARN  16:44:55,465 [request_timeout_in_ms, read_request_timeout_in_ms, range_request_timeout_in_ms, truncate_request_timeout_in_ms, write_request_timeout_in_ms] parameters have been deprecated. They have new names and/or value format; For more information, please refer to NEWS.txt
Error scrubbing BigTableReader:big(path='/private/var/folders/4d/zfjs7m7s6x5_l93k33r5k6680000gn/T/dtest-duevfyji/test/node1/data0/ks/users-7065ff30fa1311ef8b9d41e6a3a9c309/nb-1-big-Data.db'): null
java.lang.NullPointerException
	at org.apache.cassandra.utils.Interval$1.compare(Interval.java:78)
	at org.apache.cassandra.utils.Interval$1.compare(Interval.java:70)
	at org.apache.cassandra.utils.IntervalTree.buildUpdatedArray(IntervalTree.java:228)
	at org.apache.cassandra.utils.IntervalTree.update(IntervalTree.java:188)
	at org.apache.cassandra.db.lifecycle.SSTableIntervalTree.update(SSTableIntervalTree.java:85)
	at org.apache.cassandra.db.lifecycle.View$4.apply(View.java:310)
	at org.apache.cassandra.db.lifecycle.View$4.apply(View.java:305)
	at com.google.common.base.Functions$FunctionComposition.apply(Functions.java:255)
	at org.apache.cassandra.db.lifecycle.Tracker.apply(Tracker.java:167)
	at org.apache.cassandra.db.lifecycle.Tracker.apply(Tracker.java:137)
	at org.apache.cassandra.db.lifecycle.LifecycleTransaction.checkpoint(LifecycleTransaction.java:365)
	at org.apache.cassandra.db.lifecycle.LifecycleTransaction.checkpoint(LifecycleTransaction.java:346)
	at org.apache.cassandra.io.sstable.SSTableRewriter.switchWriter(SSTableRewriter.java:275)
	at org.apache.cassandra.io.sstable.SSTableRewriter.doPrepare(SSTableRewriter.java:321)
	at org.apache.cassandra.utils.concurrent.Transactional$AbstractTransactional.prepareToCommit(Transactional.java:168)
	at org.apache.cassandra.utils.concurrent.Transactional$AbstractTransactional.finish(Transactional.java:179)
	at org.apache.cassandra.io.sstable.SSTableRewriter.finish(SSTableRewriter.java:308)
	at org.apache.cassandra.io.sstable.format.SortedTableScrubber.scrub(SortedTableScrubber.java:200)
	at org.apache.cassandra.tools.StandaloneScrubber.main(StandaloneScrubber.java:173), but the regex used to check stderr is ^(WARN.*JNA link failure.*unavailable.|objc.*Class JavaLaunchHelper.*?Which one is undefined.|Failed to connect over JMX; not collecting these stats|Picked up JAVA_TOOL_OPTIONS:.*|.*parameters have been deprecated. They have new names and/or value format; For more information, please refer to NEWS.txt*|.*Error while computing token map for keyspace system_cluster_metadata with datacenter .*|\s*|\n)*$
	[<TracebackEntry /Users/maedhroz/Forks/cassandra-dtest/scrub_test.py:478>, <TracebackEntry /Users/maedhroz/Forks/cassandra-dtest/scrub_test.py:168>, <TracebackEntry /Users/maedhroz/Forks/cassandra-dtest/scrub_test.py:136>, <TracebackEntry /Users/maedhroz/Forks/cassandra-dtest/tools/assertions.py:315>]
{noformat}

Any chance that's related to this Jira?

> Improve IntervalTree build throughput
> -------------------------------------
>
>                 Key: CASSANDRA-19596
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-19596
>             Project: Apache Cassandra
>          Issue Type: Improvement
>          Components: Local/Compaction, Local/SSTable
>            Reporter: Ariel Weisberg
>            Assignee: Ariel Weisberg
>            Priority: Normal
>             Fix For: 4.0.x, 4.1.x, 5.0.x, 5.x
>
>         Attachments: IntervalTreePerfTest.java, ci_summary-cassandra-5.0-4386d87516a85ab54c5df44862a2f7fbcb5089d7.html, ci_summary-cassandra-5.0-dd23271944aa72b294054cab270c3e8f22100668.html, ci_summary-trunk-1a8b0b8957485a0cb74dd98554aef6ae9f81f0b1.html, ci_summary-trunk-b7ce75905807bde673382799f3c3bf247802bba4.html, ci_summary.html, trunk_ci_summary.html
>
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> With several terabytes of data and 8 compactors it’s possible for the compactors to spend a lot of time blocked waiting on IntervalTrees to be built.
> There is also a lot of wasted CPU because it’s updated optimistically so most of them end up being thrown away.
> This can end up being quite painful because it can block memtable flushing as well and then a single slow CFS can block unrelated CFS because the memtable post flush executor is single threaded and shared across all CFS. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org