[
https://issues.apache.org/jira/browse/CASSANDRA-3017?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13501465#comment-13501465
]
Jonathan Ellis commented on CASSANDRA-3017:
-------------------------------------------
I've been thinking about that last part and I don't have a good solution. I
*really* don't want to add the complexity of multi-part replies.
Since the goal here is to protect against malformed messages from non-cluster
members, how about this?
We pick a limit large enough to let gossip messages through (64K?) and drop
*all* messages larger than that *from non-cluster members* (according to
Gossiper's knowledge).
This doesn't help you if you're fuzz-testing Cassandra from one of your other
nodes, but hopefully you're not doing that...
I think this also would cause some dropped messages when FD announces that a
node is back up and the cluster tries to replicate stuff to it, but it doesn't
know yet that the source is a cluster member. This should be okay for reads
(since they will be small) but could cause some extra timeouts for writes.
I could see just accepting this, or adding a delay to FD recognition to keep
that as UAE for a bit longer. Or we could just wontfix.
What do you think [~brandon.williams]?
> add a Message size limit
> ------------------------
>
> Key: CASSANDRA-3017
> URL: https://issues.apache.org/jira/browse/CASSANDRA-3017
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Kirk True
> Priority: Minor
> Labels: lhf
> Attachments:
> 0001-use-the-thrift-max-message-size-for-inter-node-messa.patch,
> trunk-3017.txt
>
>
> We protect the server from allocating huge buffers for malformed message with
> the Thrift frame size (CASSANDRA-475). But we don't have similar protection
> for the inter-node Message objects.
> Adding this would be good to deal with malicious adversaries as well as a
> malfunctioning cluster participant.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
