Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Plain View
Kafka >> mail # user >> Re: How do you handle server down in producer


Copy link to this message
-
Re: How do you handle server down in producer
Not much you can do, right? Either buffer the Events in memory (for as much
memory as you have) or drop it.

IMHO if Kafka is a critical piece of your Infrastructure, it's hard to
solve the issue of what to do when it goes down. Instead, engineer your
system so your Producers and Kafka Brokers are reliable and redundant --
and well tested.

Philip

On Thu, Apr 4, 2013 at 2:30 PM, Jamie Wang <[EMAIL PROTECTED]> wrote:

> What is the best way to handle kafka server going down during a message
> send by a producer? We have seen our producer stuck waiting until kafka
> server comes back. Is there ways the producer can catch a timeout or ways
> to test if Kafka server is up so to send the message. Having the producer
> hang in memory and not do anything indefinitly is not desirable in our
> scenario.  Thanks for your help.
>
> Jamie
>

 
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB