kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Fumo, Vincent" <Vincent_F...@cable.comcast.com>
Subject Re: Metadata Request Loop?
Date Fri, 22 Apr 2016 19:15:51 GMT
Hi. I've not set that value. My producer properties are as follows :

acks=all
retries=0
bath.size=1638
linger.ms=1
buffer.memory=33554432
compression.type=gzip
client.id=sds-merdevl

I have this running on two hosts with the same config. I thought that having the same client.id
on each would just consolidate the tracking (same logical name). You don't think there is
an issue with 2 producers to the same topic?




> On Apr 22, 2016, at 3:05 PM, vinay sharma <vinsharma.tech@gmail.com> wrote:
> 
> Generally a proactive metadata refresh request is sent by producer and
> consumer every 5 minutes but this interval can be overriden with property "
> metadata.max.age.ms" which has default value 300000 i.e 5 minutes. Check if
> you have set this property very low in your producer?
> 
> On Fri, Apr 22, 2016 at 11:46 AM, Fumo, Vincent <
> Vincent_Fumo@cable.comcast.com> wrote:
> 
>> I'm testing a kafka install and using the java client. I have a topic set
>> up and it appears to work great, but after a while I noticed my log
>> starting to fill up with what appears to be some kind of loop for metadata
>> updates.
>> 
>> example::
>> 
>> 2016-04-22 15:43:55,139 DEBUG s=s-root_out  env="md"
>> [kafka-producer-network-thread | sds-merdevl]
>> org.apache.kafka.clients.Metadata: Updated cluster metadata version 6196 to
>> Cluster(nodes = [Node(0, cmp-arch-kafka-01d.cc.com, 9092)], partitions =
>> [Partition(topic = s.notifications.dev, partition = 0, leader = 0, replicas
>> = [0,], isr = [0,]])
>> 2016-04-22 15:43:55,240 DEBUG s=s-root_out  env="md"
>> [kafka-producer-network-thread | sds-merdevl]
>> org.apache.kafka.clients.Metadata: Updated cluster metadata version 6197 to
>> Cluster(nodes = [Node(0, cmp-arch-kafka-01d.cc.com, 9092)], partitions =
>> [Partition(topic = s.notifications.dev, partition = 0, leader = 0, replicas
>> = [0,], isr = [0,]])
>> 2016-04-22 15:43:55,341 DEBUG s=s-root_out  env="md"
>> [kafka-producer-network-thread | sds-merdevl]
>> org.apache.kafka.clients.Metadata: Updated cluster metadata version 6198 to
>> Cluster(nodes = [Node(0, cmp-arch-kafka-01d.cc.com, 9092)], partitions =
>> [Partition(topic = s.notifications.dev, partition = 0, leader = 0, replicas
>> = [0,], isr = [0,]])
>> 
>> etc.
>> 
>> It hasn't stopped..
>> 
>> I'm curious about what's going on here. Can anyone help?
>> 
>> 
>> 
>> 
>> 


Mime
View raw message