[jira] [Comment Edited] (CASSANDRA-6731) Requests unnecessarily redirected with CL=ONE

["Ryan McGuire (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-6731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13906293#comment-13906293 ] 

Ryan McGuire edited comment on CASSANDRA-6731 at 2/19/14 11:48 PM:
-------------------------------------------------------------------

[~thobbs] yep, setting speculative_retry='NONE' dropped the connections to other nodes entirely. FWIW, I haven't seen the performance problem that [~arisp] originally saw (he [reported 50% slowdown here|http://stackoverflow.com/questions/21807422/cassandra-reads-slow-with-multiple-nodes].) SR does seem to usually decrease best-case performance a bit, but drastically increase performance in the worst-case, so I'm happy as it stands.


was (Author: enigmacurry):
[~thobbs] yep, setting speculative_retry='NONE' dropped the connections to other nodes entirely. FWIW, I haven't seen the performance problem that [~arisp] originally saw. SR does seem to usually decrease best-case performance a bit, but drastically increase performance in the worst-case, so I'm happy as it stands.

> Requests unnecessarily redirected with CL=ONE
> ---------------------------------------------
>
>                 Key: CASSANDRA-6731
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6731
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>         Environment: Linux
>            Reporter: Aris Prassinos
>            Assignee: Tyler Hobbs
>            Priority: Minor
>             Fix For: 2.0.6
>
>         Attachments: 6731.traces.txt, ticket_6731.py
>
>
> Three-node cluster with RF=3. All data currently in sync.
> Network topology strategy. Each node defined to be on a different rack.
> endpoint_snitch: PropertyFileSnitch
> dynamic_snitch_update_interval_in_ms: 100
> dynamic_snitch_reset_interval_in_ms: 600000
> dynamic_snitch_badness_threshold: 0
> All tables defined with read_repair_chance=0
> From cqlsh when querying with Consistency=ONE tracing shows that requests get frequently redirected to other nodes even though the local node has the data.
> There is no other activity on the cluster besides my test queries.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)