[
https://issues.apache.org/jira/browse/CASSANDRA-9318?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14535511#comment-14535511
]
Ariel Weisberg commented on CASSANDRA-9318:
-------------------------------------------
OK, so how does this gel with availability? Does this mean C* supports partial
failures where a requests involving that node will get OE, but other requests
are not impeded? With vnodes this will mean total unavalability.
I'm glad OE is in the protocol. There are use cases where upstream isn't going
to want the DB to silently pretend it could accept the load and wants feedback
instead.
Benedict can you remind me about what test your thinking of? I think it may not
have been 8670.
> Bound the number of in-flight requests at the coordinator
> ---------------------------------------------------------
>
> Key: CASSANDRA-9318
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9318
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Ariel Weisberg
> Assignee: Ariel Weisberg
> Fix For: 2.1.x
>
>
> It's possible to somewhat bound the amount of load accepted into the cluster
> by bounding the number of in-flight requests and request bytes.
> An implementation might do something like track the number of outstanding
> bytes and requests and if it reaches a high watermark disable read on client
> connections until it goes back below some low watermark.
> Need to make sure that disabling read on the client connection won't
> introduce other issues.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
