Do you have a simple test that can reproduce this issue?
Thanks,
Jun
On Thu, Oct 30, 2014 at 8:34 PM, Bhavesh Mistry
wrote:
> HI Jun,
>
> Consumer Connector is not closed because I can see the ConsumerFetcher
> Thread alive but Blocked on *put* and hasNext() is blocked on *take*.
> This is wha
Hey Guys,
If at the beginning, I configure the producer to not use compression
and produce a number of message to a topic. Then later, I reconfigure the
producer to use compression and produce another batch of messages to the
same topic. Can a consumer correctly read all the messages in the t
Hi Zuoning,
Yes, the same topic's partitions can hold both compressed and uncompressed
data, and the consumer is able to read them in mixed mode.
Guozhang
On Fri, Oct 31, 2014 at 8:40 AM, Zuoning Yin wrote:
> Hey Guys,
> If at the beginning, I configure the producer to not use compressio
Thanks so much for the reply, Guozhang!
On Fri, Oct 31, 2014 at 12:06 PM, Guozhang Wang wrote:
> Hi Zuoning,
>
> Yes, the same topic's partitions can hold both compressed and uncompressed
> data, and the consumer is able to read them in mixed mode.
>
> Guozhang
>
> On Fri, Oct 31, 2014 at 8:40 A
Hi Guozhang,
Just want to double check: does this have some requirement on the Kafka
version (we are using 0.8.0) or the kafka client (we are using librdkafka)?
Thanks,
--Zuoning
On Fri, Oct 31, 2014 at 12:09 PM, Zuoning Yin wrote:
> Thanks so much for the reply, Guozhang!
>
> On Fri, Oct 3
The server.log should be the broker logs.
Guozhang
On Thu, Oct 30, 2014 at 11:32 AM, Sameer Yami wrote:
> I do not see any other logs besides these.
> I checked the log4j.properties and where the logs are getting stored.
> The only other logs (apart from the controller logs) in the directory ar
In your instance if you have four JVMs (i.e., consumer processes), six
threads per consumer process and 12 partitions, then each thread would
only get one partition but the first two processes will get all the
partitions and the last two processes would be idle. We could tweak
the assignment strate
To circle back on this thread. The patch in kafka-1482 is almost ready. To
make the mbean names more meaningful and easier to parse, the patch will
use explicit key/value pairs in the mbean name for things like clientId and
topic, and will get rid of the quotes.
So, instead of
"kafka.server":ty
That sounds good, although is that the only change (sorry I have not
done a careful review of that patch and would like to before it gets
checked in).
On Fri, Oct 31, 2014 at 10:42:13AM -0700, Jun Rao wrote:
> To circle back on this thread. The patch in kafka-1482 is almost ready. To
> make the mb
Yes, all changes are related to metric names. Feel free to review the patch.
Thanks,
Jun
On Fri, Oct 31, 2014 at 10:48 AM, Joel Koshy wrote:
> That sounds good, although is that the only change (sorry I have not
> done a careful review of that patch and would like to before it gets
> checked i
Note: https://issues.apache.org/jira/browse/KAFKA-1481 (not 1482, just so
there is no confusion)
Otis
--
Monitoring * Alerting * Anomaly Detection * Centralized Log Management
Solr & Elasticsearch Support * http://sematext.com/
On Fri, Oct 31, 2014 at 2:10 PM, Jun Rao wrote:
> Yes, all changes
Hi All,
I’m new to Kafka, please direct me to the right path if it is a duplicate
question.
Basically I deployed Kafka to a 4 machine cluster, what if the whole cluster
went down, does kafka provide any backup/restore mechanism? Please advise.
Thanks!
Yingkai
Hi
I'm trying to upgrade a 0.7 kafka cluster, but I'm getting an error:
I created the file migrationToolConsumer.properties with just theinfo to
connect to the old cluster:
_
zookeeper.connect=zookeeper01:2181,zookeeper02:2181,zookeeper03:2181
group.id=migration.prod
__
Then the produce
Well... I used strace and I found it was looking for some classes in a
wrong path. I fixed most of them, but there's one that isn't anywhere,
neither the new nor the old kafka servers:
$ strace -o logfile -s 1000 -f kafka/bin/kafka-run-class.sh
kafka.tools.KafkaMigrationTool --kafka.07.jar
migrati
The producer configuration should list the kafka brokers, not the zookeeper
quorum.
See here: http://kafka.apache.org/documentation.html#producerconfigs
(and send my regards to Alex Gorbachev ;)
Gwen
On Fri, Oct 31, 2014 at 8:05 PM, Tomas Nunez wrote:
> Hi
>
> I'm trying to upgrade a 0.7 kaf
This is part of Scala, so it should be in the scala-library-...jar
On Fri, Oct 31, 2014 at 8:26 PM, Tomas Nunez wrote:
> Well... I used strace and I found it was looking for some classes in a
> wrong path. I fixed most of them, but there's one that isn't anywhere,
> neither the new nor the old
Thanks for pointing the error. I fixed it, but I'm still getting the same
error. The content of migrationToolProducer.properties is now:
_
metadata.broker.list=kafka_new01:9092,kafka_new02:9092,kafka_new03:9092,kafka_new04:9092,kafka_new05:9092,
group.id=migration.prod
_
And the error stil
The file should be your Scala jar.
You should have it in your Kafka lib directory.
Do a "jar -t" to make sure it actually contains the Tuple class.
Gwen
On Fri, Oct 31, 2014 at 8:49 PM, Tomas Nunez wrote:
> Thanks for pointing the error. I fixed it, but I'm still getting the same
> error. The c
Hi Jun,
Here is code base:
https://github.com/bmistry13/kafka-trunk-producer/blob/master/KafkaConsumerWithDelay.java
Please let me know if you can help me determine the root cause. Why
there is illegal state and blocking ?
Thanks,
Bhavesh
On Fri, Oct 31, 2014 at 8:33 AM, Jun Rao wrote:
>
19 matches
Mail list logo
