Re: erlang client for 0.8?

If you are only looking for a producer, I maintain one that is used in production => http://github.com/helpshift/ekaf kafboy is http gateway over ekaf that handles ~100 million events pushed per day. ( it uses the awesome cowboy as the underlying http server) => http://github.com/helpshift/kafboy

Re: How try out the new producer?

Good to know. Thank you, Jun. On Fri, Aug 15, 2014 at 4:03 PM, Jun Rao wrote: > You can use the "-x" option to skip signing. This is described in README. > > Thanks, > > Jun > > > On Fri, Aug 15, 2014 at 11:23 AM, cac...@gmail.com > wrote: > > > I believe I downloaded from trunk and compiled

Re: Most common kafka client comsumer implementations?

Neha, I believe option 2 is more like the high level consumer, no? High level consumer doesn't really guarantee processing of the message. E.g. I take 1000 messages feed them off to a bunch of threads and "hope and pray" they get processed. Option 3 is something I haven't seen in the kafka client w

Re: How try out the new producer?

You can use the "-x" option to skip signing. This is described in README. Thanks, Jun On Fri, Aug 15, 2014 at 11:23 AM, cac...@gmail.com wrote: > I believe I downloaded from trunk and compiled a jar from that. The hardest > part of that seemed to be configuring gradle to sign the jar having n

Re: Strange topic-corruption issue?

What's in there seems to be still the output for deep iteration. For shallow iteration, the compression codec for each message should be Snappy. Thanks, Jun On Fri, Aug 15, 2014 at 5:27 AM, Steve Miller wrote: >Oh, yeah, sorry about that. I threw a copy of that up at: > > https:/

Re: How try out the new producer?

In case this helps anyone else, this seemed to work for me (not bothering with signing). ./gradlew clients:install Then use/copy jar and pom from ~/.m2/repository/org/apache/kafka/kafka-clients/0.8.2-SNAPSHOT On Fri, Aug 15, 2014 at 11:23 AM, cac...@gmail.com wrote: > I believe I downloaded fr

Re: kafka 0.8.1.1 log.retention.minutes NOT being honored

UP! Can someone add the log.retention.hours in the documentation? And is the retention.hours or retention.minutes available for the by-topic configuration? François Langelier Étudiant en génie Logiciel - École de Technologie Supérieure Capitaine Club Capra

Re: How try out the new producer?

Thanks Neha and Christian for the responses. I'll give it a go. On Fri, Aug 15, 2014 at 11:22 AM, Neha Narkhede wrote: > The most stable version of the new producer is in trunk. To get it you will > have to build trunk and use it from the clients jar. > > > On Fri, Aug 15, 2014 at 11:00 AM, Ro

Re: How try out the new producer?

I believe I downloaded from trunk and compiled a jar from that. The hardest part of that seemed to be configuring gradle to sign the jar having never done it before. Christian On Fri, Aug 15, 2014 at 11:00 AM, Roger Hoover wrote: > Hi, > > I want to try out the new producer api > (org.apache.k

Re: How try out the new producer?

The most stable version of the new producer is in trunk. To get it you will have to build trunk and use it from the clients jar. On Fri, Aug 15, 2014 at 11:00 AM, Roger Hoover wrote: > Hi, > > I want to try out the new producer api > (org.apache.kafka.clients.producer.KafkaProducer) but found t

How try out the new producer?

Hi, I want to try out the new producer api (org.apache.kafka.clients.producer.KafkaProducer) but found that it's not in the published jar. What's the best way to get it? Build from source from the 0.8.1.1 tag? Any flags I need to set to include the new producer in the jar? Thanks, Roger

Re: JMSXGroupID clustered message grouping

I am not familiar with "JMSXGroupID", but it sounds like you could just use the Producer API which allows your code to choose the partition to which a given message is sent -- perform some modulo math on your GroupID, given your number of partitions ("partitioner.class"). And since only 1 Consum

JMSXGroupID clustered message grouping

Hi, When using multiple partitions and consumers, can Kafka ensure that JMS messages with the same JMSXGroupID are handled by the same consumer? Or does this have to be implemented when writing the Producers and Consumers? For reference, this is in comparison to the clustered message grouping fea

RE: Consumer sensitive expiration of topic

Hi, Thanks for the answer. The topics themselves won't be shortlived (as their consumers are supposed to stay there), the messages in them will. What i'm trying to achieve is something similar to this: Producers > Processor A0 > Processor A1 > ... > Consumer

Re: Strange topic-corruption issue?

Oh, yeah, sorry about that. I threw a copy of that up at: https://newbie.idrathernotsay.com/full.txt.gz (you'll get a cert error, for the four times I put something on the home website each year, I didn't feel like getting a real cert (-: ). If that doesn't work I'm sure I can fi