[
https://issues.apache.org/jira/browse/CASSANDRA-9318?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14604726#comment-14604726
]
Jonathan Ellis commented on CASSANDRA-9318:
-------------------------------------------
We do not drop hints on the floor. We abort a write with UAE if we have too
many hints in progress, but that is okay because we've told the client to
expect it. Here's a quick summary:
# If you send timedoutexception to a client, then you have to write a hint
# If you send UAE, then you do not have to write a hint
This is equivalent to
# If you send the write to any replicas, you must send it to all (via hint if
necessary)
# If you send UAE, you must do so before sending to any replicas
> 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: 2.2.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)
