[jira] [Commented] (CASSANDRA-5683) Write timeout in multi-dc environment

["sankalp kohli (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-5683?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13784398#comment-13784398 ] 

sankalp kohli commented on CASSANDRA-5683:
------------------------------------------

Timeouts will differ based on a combination of many things like operation time(read, write), consistency level(multi-DC or not) and whether it is a batch or single operation. 
Also you will want different timeouts for  multi-DC hops vs within the same DC. 
Instead of having so many properties for each combination, we should let client decide the timeout and should propagate that timeout. This is because client knows which combination it is using. We might also allow passing separate timeouts for within DC and multi-DC reads or writes. 
This will be very useful for reads as there is no point working on timed out queries. 

> Write timeout in multi-dc environment 
> --------------------------------------
>
>                 Key: CASSANDRA-5683
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5683
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 1.1.9
>         Environment: apache cassandra 1.1.9
>            Reporter: Boole Guo
>              Labels: write
>
> When writing in mutil-dc environment, there are many timeout exception. As I know, this version have do good to reduce network bandwidth. Can we assign rpc timeout when writing like consistenylevel?



--
This message was sent by Atlassian JIRA
(v6.1#6144)