[jira] [Updated] (CASSANDRA-14139) Acquire read lock before accessing CompactionStrategyManager fields
[
https://issues.apache.org/jira/browse/CASSANDRA-14139?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Paulo Motta updated CASSANDRA-14139:
Resolution: Fixed
Fix Version/s: 4.0
3.11.2
Reproduced In: (was: 3.11.2)
Status: Resolved (was: Ready to Commit)
Committed as {{fe0ee85c71faada0acb48a65f249575c65bf0972}} to cassandra-3.11 and
merged up to master. Thanks!
> Acquire read lock before accessing CompactionStrategyManager fields
> ---
>
> Key: CASSANDRA-14139
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14139
> Project: Cassandra
> Issue Type: Bug
>Reporter: Paulo Motta
>Assignee: Paulo Motta
> Fix For: 3.11.2, 4.0
>
> Attachments: 3.11-14139-dtest.png, 3.11-14139-testall.png,
> trunk-14139-dtest.png, trunk-14139-testall.png
>
>
> There are a few methods in {{CompactionStrategyManager}} accessing the
> repaired/unrepaired compaction strategy lists without using the read lock,
> what could cause issues like the one below:
> {noformat}
> ERROR [CompactionExecutor:1] 2017-12-22 12:17:12,320 CassandraDaemon.java:141
> - Exception in thread Thread[CompactionExecutor:1,5,main]
> java.lang.IndexOutOfBoundsException: Index: 0, Size: 1
> at java.util.ArrayList.rangeCheck(ArrayList.java:657)
> at java.util.ArrayList.get(ArrayList.java:433)
> at
> org.apache.cassandra.db.compaction.CompactionStrategyManager.supportsEarlyOpen(CompactionStrategyManager.java:1262)
> at
> org.apache.cassandra.db.ColumnFamilyStore.supportsEarlyOpen(ColumnFamilyStore.java:558)
> at
> org.apache.cassandra.io.sstable.SSTableRewriter.construct(SSTableRewriter.java:119)
> at
> org.apache.cassandra.db.compaction.writers.CompactionAwareWriter.(CompactionAwareWriter.java:91)
> at
> org.apache.cassandra.db.compaction.writers.DefaultCompactionWriter.(DefaultCompactionWriter.java:57)
> at
> org.apache.cassandra.db.compaction.CompactionTask.getCompactionAwareWriter(CompactionTask.java:293)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:200)
> at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:90)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:101)
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:310)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:81)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-14139) Acquire read lock before accessing CompactionStrategyManager fields
[
https://issues.apache.org/jira/browse/CASSANDRA-14139?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marcus Eriksson updated CASSANDRA-14139:
Status: Ready to Commit (was: Patch Available)
> Acquire read lock before accessing CompactionStrategyManager fields
> ---
>
> Key: CASSANDRA-14139
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14139
> Project: Cassandra
> Issue Type: Bug
>Reporter: Paulo Motta
>Assignee: Paulo Motta
> Attachments: 3.11-14139-dtest.png, 3.11-14139-testall.png,
> trunk-14139-dtest.png, trunk-14139-testall.png
>
>
> There are a few methods in {{CompactionStrategyManager}} accessing the
> repaired/unrepaired compaction strategy lists without using the read lock,
> what could cause issues like the one below:
> {noformat}
> ERROR [CompactionExecutor:1] 2017-12-22 12:17:12,320 CassandraDaemon.java:141
> - Exception in thread Thread[CompactionExecutor:1,5,main]
> java.lang.IndexOutOfBoundsException: Index: 0, Size: 1
> at java.util.ArrayList.rangeCheck(ArrayList.java:657)
> at java.util.ArrayList.get(ArrayList.java:433)
> at
> org.apache.cassandra.db.compaction.CompactionStrategyManager.supportsEarlyOpen(CompactionStrategyManager.java:1262)
> at
> org.apache.cassandra.db.ColumnFamilyStore.supportsEarlyOpen(ColumnFamilyStore.java:558)
> at
> org.apache.cassandra.io.sstable.SSTableRewriter.construct(SSTableRewriter.java:119)
> at
> org.apache.cassandra.db.compaction.writers.CompactionAwareWriter.(CompactionAwareWriter.java:91)
> at
> org.apache.cassandra.db.compaction.writers.DefaultCompactionWriter.(DefaultCompactionWriter.java:57)
> at
> org.apache.cassandra.db.compaction.CompactionTask.getCompactionAwareWriter(CompactionTask.java:293)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:200)
> at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:90)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:101)
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:310)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:81)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-14139) Acquire read lock before accessing CompactionStrategyManager fields
[
https://issues.apache.org/jira/browse/CASSANDRA-14139?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Paulo Motta updated CASSANDRA-14139:
Reviewer: Marcus Eriksson
> Acquire read lock before accessing CompactionStrategyManager fields
> ---
>
> Key: CASSANDRA-14139
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14139
> Project: Cassandra
> Issue Type: Bug
>Reporter: Paulo Motta
>Assignee: Paulo Motta
> Attachments: 3.11-14139-dtest.png, 3.11-14139-testall.png,
> trunk-14139-dtest.png, trunk-14139-testall.png
>
>
> There are a few methods in {{CompactionStrategyManager}} accessing the
> repaired/unrepaired compaction strategy lists without using the read lock,
> what could cause issues like the one below:
> {noformat}
> ERROR [CompactionExecutor:1] 2017-12-22 12:17:12,320 CassandraDaemon.java:141
> - Exception in thread Thread[CompactionExecutor:1,5,main]
> java.lang.IndexOutOfBoundsException: Index: 0, Size: 1
> at java.util.ArrayList.rangeCheck(ArrayList.java:657)
> at java.util.ArrayList.get(ArrayList.java:433)
> at
> org.apache.cassandra.db.compaction.CompactionStrategyManager.supportsEarlyOpen(CompactionStrategyManager.java:1262)
> at
> org.apache.cassandra.db.ColumnFamilyStore.supportsEarlyOpen(ColumnFamilyStore.java:558)
> at
> org.apache.cassandra.io.sstable.SSTableRewriter.construct(SSTableRewriter.java:119)
> at
> org.apache.cassandra.db.compaction.writers.CompactionAwareWriter.(CompactionAwareWriter.java:91)
> at
> org.apache.cassandra.db.compaction.writers.DefaultCompactionWriter.(DefaultCompactionWriter.java:57)
> at
> org.apache.cassandra.db.compaction.CompactionTask.getCompactionAwareWriter(CompactionTask.java:293)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:200)
> at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:90)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:101)
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:310)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:81)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-14139) Acquire read lock before accessing CompactionStrategyManager fields
[
https://issues.apache.org/jira/browse/CASSANDRA-14139?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Paulo Motta updated CASSANDRA-14139:
Status: Patch Available (was: Open)
It seems we missed grabbing the read lock for some operations on
CASSANDRA-13948, what can cause IndexOutOfBounds/NullPointers such the ones
above during compaction strategy reload, so this patch basically gets the lock
on the few methods were not doing it.
CI looks good. Trunk patch is slightly different because there are some
additional methods due to CASSANDRA-9143. Can you take a look [~krummas]?
Thanks!
||3.11||trunk||
|[branch|https://github.com/apache/cassandra/compare/cassandra-3.11...pauloricardomg:3.11-14139]|[branch|https://github.com/apache/cassandra/compare/trunk...pauloricardomg:trunk-14139]|
|[testall|https://issues.apache.org/jira/secure/attachment/12904036/3.11-14139-testall.png]|[testall|https://issues.apache.org/jira/secure/attachment/12904038/trunk-14139-testall.png]|
|[dtest|https://issues.apache.org/jira/secure/attachment/12904035/3.11-14139-dtest.png]|[dtest|https://issues.apache.org/jira/secure/attachment/12904037/trunk-14139-dtest.png]|
> Acquire read lock before accessing CompactionStrategyManager fields
> ---
>
> Key: CASSANDRA-14139
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14139
> Project: Cassandra
> Issue Type: Bug
>Reporter: Paulo Motta
>Assignee: Paulo Motta
> Attachments: 3.11-14139-dtest.png, 3.11-14139-testall.png,
> trunk-14139-dtest.png, trunk-14139-testall.png
>
>
> There are a few methods in {{CompactionStrategyManager}} accessing the
> repaired/unrepaired compaction strategy lists without using the read lock,
> what could cause issues like the one below:
> {noformat}
> ERROR [CompactionExecutor:1] 2017-12-22 12:17:12,320 CassandraDaemon.java:141
> - Exception in thread Thread[CompactionExecutor:1,5,main]
> java.lang.IndexOutOfBoundsException: Index: 0, Size: 1
> at java.util.ArrayList.rangeCheck(ArrayList.java:657)
> at java.util.ArrayList.get(ArrayList.java:433)
> at
> org.apache.cassandra.db.compaction.CompactionStrategyManager.supportsEarlyOpen(CompactionStrategyManager.java:1262)
> at
> org.apache.cassandra.db.ColumnFamilyStore.supportsEarlyOpen(ColumnFamilyStore.java:558)
> at
> org.apache.cassandra.io.sstable.SSTableRewriter.construct(SSTableRewriter.java:119)
> at
> org.apache.cassandra.db.compaction.writers.CompactionAwareWriter.(CompactionAwareWriter.java:91)
> at
> org.apache.cassandra.db.compaction.writers.DefaultCompactionWriter.(DefaultCompactionWriter.java:57)
> at
> org.apache.cassandra.db.compaction.CompactionTask.getCompactionAwareWriter(CompactionTask.java:293)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:200)
> at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:90)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:101)
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:310)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:81)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Updated] (CASSANDRA-14139) Acquire read lock before accessing CompactionStrategyManager fields
[
https://issues.apache.org/jira/browse/CASSANDRA-14139?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Paulo Motta updated CASSANDRA-14139:
Summary: Acquire read lock before accessing CompactionStrategyManager
fields (was: Avoid races during compaction strategy reload)
> Acquire read lock before accessing CompactionStrategyManager fields
> ---
>
> Key: CASSANDRA-14139
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14139
> Project: Cassandra
> Issue Type: Bug
>Reporter: Paulo Motta
>Assignee: Paulo Motta
> Attachments: 3.11-14139-dtest.png, 3.11-14139-testall.png,
> trunk-14139-dtest.png, trunk-14139-testall.png
>
>
> There are a few methods in {{CompactionStrategyManager}} accessing the
> repaired/unrepaired compaction strategy lists without using the read lock,
> what could cause issues like the one below:
> {noformat}
> ERROR [CompactionExecutor:1] 2017-12-22 12:17:12,320 CassandraDaemon.java:141
> - Exception in thread Thread[CompactionExecutor:1,5,main]
> java.lang.IndexOutOfBoundsException: Index: 0, Size: 1
> at java.util.ArrayList.rangeCheck(ArrayList.java:657)
> at java.util.ArrayList.get(ArrayList.java:433)
> at
> org.apache.cassandra.db.compaction.CompactionStrategyManager.supportsEarlyOpen(CompactionStrategyManager.java:1262)
> at
> org.apache.cassandra.db.ColumnFamilyStore.supportsEarlyOpen(ColumnFamilyStore.java:558)
> at
> org.apache.cassandra.io.sstable.SSTableRewriter.construct(SSTableRewriter.java:119)
> at
> org.apache.cassandra.db.compaction.writers.CompactionAwareWriter.(CompactionAwareWriter.java:91)
> at
> org.apache.cassandra.db.compaction.writers.DefaultCompactionWriter.(DefaultCompactionWriter.java:57)
> at
> org.apache.cassandra.db.compaction.CompactionTask.getCompactionAwareWriter(CompactionTask.java:293)
> at
> org.apache.cassandra.db.compaction.CompactionTask.runMayThrow(CompactionTask.java:200)
> at org.apache.cassandra.utils.WrappedRunnable.run(WrappedRunnable.java:28)
> at
> org.apache.cassandra.db.compaction.CompactionTask.executeInternal(CompactionTask.java:90)
> at
> org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(AbstractCompactionTask.java:101)
> at
> org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run(CompactionManager.java:310)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at
> org.apache.cassandra.concurrent.NamedThreadFactory.lambda$threadLocalDeallocator$0(NamedThreadFactory.java:81)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
