You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@kafka.apache.org by Amar <am...@gmail.com> on 2015/08/26 22:36:29 UTC

Reg Leader re-election when leader crashes

Hi,

I have configured 3 brokers and 3 zookeepers in different unix boxes. All
are receiving message successfully.

1. Now leader broker got crashed and when I try to publish 10 message. some
of them are getting failed with
org.apache.kafka.common.errors.TimeoutException: Failed to update metadata
after 30000 ms.
Can you please advise

2.  two broker including leader broker got crashed. I ran below command and
found that some offline partiitions. Please advise why they are not getting
re-assigned to third broker ?

bin/kafka-topics.sh --describe --zookeeper localhost:2181 --topic
RTL_RESERVATION_UPDATES --unavailable-partitions
Regards
Amar

Re: Reg Leader re-election when leader crashes

Posted by Ashish Singh <as...@cloudera.com>.
Hello Amar,

Few questions.

1. What is the version of Kafka you are using?
2. What is the replication factor of your topic?
3. What is your min.insync.replicas set to?
4. Are you using new producer? Are you producing in sync or async mode?

On Wed, Aug 26, 2015 at 1:36 PM, Amar <am...@gmail.com> wrote:

> Hi,
>
> I have configured 3 brokers and 3 zookeepers in different unix boxes. All
> are receiving message successfully.
>
> 1. Now leader broker got crashed and when I try to publish 10 message. some
> of them are getting failed with
> org.apache.kafka.common.errors.TimeoutException: Failed to update metadata
> after 30000 ms.
> Can you please advise
>
> 2.  two broker including leader broker got crashed. I ran below command and
> found that some offline partiitions. Please advise why they are not getting
> re-assigned to third broker ?
>
> bin/kafka-topics.sh --describe --zookeeper localhost:2181 --topic
> RTL_RESERVATION_UPDATES --unavailable-partitions
> Regards
> Amar
>



-- 

Regards,
Ashish