[
https://issues.apache.org/jira/browse/CASSANDRA-14057?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16257389#comment-16257389
]
Jason Brown commented on CASSANDRA-14057:
-----------------------------------------
Damnit, shame on me for not backporting that fix (or just fixing it outside of
CASSANDRA-8457).
+1, [~slebresne], and thanks for taking the time to make sure we do this
correctly.
> The size of a byte is not 2
> ---------------------------
>
> Key: CASSANDRA-14057
> URL: https://issues.apache.org/jira/browse/CASSANDRA-14057
> Project: Cassandra
> Issue Type: Bug
> Components: Streaming and Messaging
> Reporter: Sylvain Lebresne
> Assignee: Sylvain Lebresne
> Priority: Trivial
> Fix For: 3.0.16, 3.11.2
>
>
> {{DataLimits}} serializer uses
> {{TypeSizes.sizeof((byte)limits.kind().ordinal())}} in it's
> {{serializedSize}} method, but {{TypeSizes}} does not (on 3.0/3.11) have a
> {{sizeof(byte)}} override, so {{sizeof(short)}} is picked and it returns 2,
> which is wrong.
> This is actually fixed on trunk, [~jasobrown] committed the fix as part of
> CASSANDRA-8457, but it wasn't committed in 3.0/3.11 and that still feel dodgy
> to leave as is.
> To be clear, I don't think it's a problem as of now: it does break the size
> computed for read commands, and thus the payload size of such message, but
> said payload size is not use (for read requests that is).
> Still, not reason to leave something wrong like this and risk a future bug
> when the fix is trivial.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
