[
https://issues.apache.org/jira/browse/CASSANDRA-15405?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sam Tunnicliffe updated CASSANDRA-15405:
----------------------------------------
Status: Changes Suggested (was: Review In Progress)
Mostly LGTM.
You could short circuit {{maybeTrimLiveCells}} where {{maxLiveCells >
cells.size()}}, which should be the more common case I would think.
Also, there are a handful of test failures due to {{command}} being null.
Mostly these are down to the test itself, but there is a legit case where a
null {{ReadCommand}} is passed, in {{LegacyBatchlogMigrator}}.
I've pushed a patch which fixes these
[here|https://github.com/beobal/cassandra/commit/18f8fe1f2b8a993efbc5fdb6bf99c2e90e7348be]
> Mixed mode reads on compact storage tables can return incomplete results
> ------------------------------------------------------------------------
>
> Key: CASSANDRA-15405
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15405
> Project: Cassandra
> Issue Type: Bug
> Components: Legacy/Core
> Reporter: Marcus Eriksson
> Assignee: Marcus Eriksson
> Priority: Normal
> Fix For: 3.0.x, 3.11.x
>
>
> In mixed mode (2.1/3.0), when coordinating a read on a 2.1 node, reading data
> from 3.0 nodes, we [incorrectly
> trim|https://github.com/apache/cassandra/blob/53f604dc1789a800dbcbc3c8aee77f8f36b8b5db/src/java/org/apache/cassandra/db/LegacyLayout.java#L529]
> the result (if it has tombstones) when preparing it for the 2.1 node, this
> is then [interpreted by the 2.1
> node|https://github.com/apache/cassandra/blob/cassandra-2.1/src/java/org/apache/cassandra/service/pager/AbstractQueryPager.java#L110]
> as the pager has been exhausted.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
