[
https://issues.apache.org/jira/browse/CASSANDRA-17401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18045743#comment-18045743
]
Alex Petrov commented on CASSANDRA-17401:
-----------------------------------------
I am personally not a big fan of adding version-dependent logic, since it
complicates maintenance and testing substantially. I think we do need to
replace invalidate with invalidateAll regardless of this ticket, since this is
a race, too. Another thing I'd do is I'd check if removal listener is
sufficient to clean up the prep statements table, and avoid explicitly calling
{{removePreparedStatement}}.
Another approach is to evict only if we see that one of the statements is
cached but not the other (i.e. {{ (a == null && b != null) && (b == null && a
!= null) }}), which is going to substantially reduce the window of race
possibility. This, combined with what I wrote earlier, namely
bq. Another addition to the patch I would do is adding an if statement when
retrieving {{cachedWithKeyspace}} after {{cachedWithoutKeyspace}}, and only
going for a second get if hashes are actually different.
should be sufficient to trigger only a handful of spurious re-prepares. WDYT?
> Race condition in QueryProcessor causes just prepared statement not to be in
> the prepared statements cache
> ----------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-17401
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17401
> Project: Apache Cassandra
> Issue Type: Bug
> Components: Messaging/Client
> Reporter: Ivan Senic
> Assignee: Jaydeepkumar Chovatia
> Priority: Normal
> Time Spent: 10m
> Remaining Estimate: 0h
>
> The changes in the
> [QueryProcessor#prepare|https://github.com/apache/cassandra/blame/cassandra-4.0.2/src/java/org/apache/cassandra/cql3/QueryProcessor.java#L575-L638]
> method that were introduced in versions *4.0.2* and *3.11.12* can cause a
> race condition between two threads trying to concurrently prepare the same
> statement. This race condition can cause removing of a prepared statement
> from the cache, after one of the threads has received the result of the
> prepare and eventually uses MD5Digest to call
> [QueryProcessor#getPrepared|https://github.com/apache/cassandra/blame/cassandra-4.0.2/src/java/org/apache/cassandra/cql3/QueryProcessor.java#L212-L215].
> The race condition looks like this:
> * Thread1 enters _prepare_ method and resolves _safeToReturnCached_ as false
> * Thread1 executes eviction of hashes
> * Thread2 enters _prepare_ method and resolves _safeToReturnCached_ as false
> * Thread1 prepares the statement and caches it
> * Thread1 returns the result of the prepare
> * Thread2 executes eviction of hashes
> * Thread1 tries to execute the prepared statement with the received
> MD5Digest, but statement is not in the cache as it was evicted by Thread2
> I tried to reproduce this by using a Java driver, but hitting this case from
> a client side is highly unlikely and I can not simulate the needed race
> condition. However, we can easily reproduce this in Stargate (details
> [here|https://github.com/stargate/stargate/pull/1647]), as it's closer to
> QueryProcessor.
> Reproducing this in a unit test is fairly easy. I am happy to showcase this
> if needed.
> Note that the issue can occur only when safeToReturnCached is resolved as
> false.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
