Re: ISR churn

2017-03-23 Thread Radu Radutiu
ks for you, I'd like to try it in our staging clusters. > > > > Thanks, > > > > Marcos Juarez > > > > > > > > > > > > On Wed, Mar 22, 2017 at 11:55 AM, Jun MA <mj.saber1...@gmail.com > <mailto:mj.saber1...@gmail.com>> wrote: >

Re: ISR churn

2017-03-22 Thread David Garcia
Sure…there are two types of purgatories: Consumer and Producer Consumer purgatory (for a partition leader) is a queue for pending requests for data (i.e. polling by some client for the respective partition). It’s basically a waiting area for poll requests. Generally speaking, the more

Re: ISR churn

2017-03-22 Thread Jun MA
Hi David, I checked our cluster, the producer purgatory size is under 3 mostly. But I’m not quite understand this metrics, could you please explain it a little bit? Thanks, Jun > On Mar 22, 2017, at 3:07 PM, David Garcia wrote: > > producer purgatory size

Re: ISR churn

2017-03-22 Thread James Cheng
55 AM, Jun MA <mj.saber1...@gmail.com > <mailto:mj.saber1...@gmail.com>> wrote: > I have similar issue with our cluster. We don’t know the root cause but we > have some interesting observation. > > 1. We do see correlation between ISR churn and fetcher connection > cl

Re: ISR churn

2017-03-22 Thread David Garcia
2, 2017 at 11:55 AM, Jun MA <mj.saber1...@gmail.com <mailto:mj.saber1...@gmail.com>> wrote: > I have similar issue with our cluster. We don’t know the root cause but we have some interesting observation. > > 1. We do see correlation between ISR churn and fetche

Re: ISR churn

2017-03-22 Thread Jun MA
> > Thanks, > > Marcos Juarez > > > > > > On Wed, Mar 22, 2017 at 11:55 AM, Jun MA <mj.saber1...@gmail.com > <mailto:mj.saber1...@gmail.com>> wrote: > I have similar issue with our cluster. We don’t know the root cause but we > have some

Re: ISR churn

2017-03-22 Thread Jeff Widman
w > broker? If that works for you, I'd like to try it in our staging clusters. > > Thanks, > > Marcos Juarez > > > > > > On Wed, Mar 22, 2017 at 11:55 AM, Jun MA <mj.saber1...@gmail.com> wrote: > >> I have similar issue with our cluster. We don’t know

Re: ISR churn

2017-03-22 Thread Marcos Juarez
se but we > have some interesting observation. > > 1. We do see correlation between ISR churn and fetcher connection > close/create. > > 2. We’ve tried to add a broker which doesn’t have any partitions on it > dedicate to the controller (rolling restart existing brokers and failover

Re: ISR churn

2017-03-22 Thread Manikumar
On Wed, Mar 22, 2017 at 7:21 PM, Radu Radutiu <rradu...@gmail.com> wrote: > Hello, > > Does anyone know how I can debug high ISR churn on the kafka leader on a > cluster without traffic? I have 2 topics on a 4 node cluster (replica 4 > and replica 3) and both show constant

ISR churn

2017-03-22 Thread Radu Radutiu
Hello, Does anyone know how I can debug high ISR churn on the kafka leader on a cluster without traffic? I have 2 topics on a 4 node cluster (replica 4 and replica 3) and both show constant changes of the number of insync replicas: [2017-03-22 15:30:10,945] INFO Partition [__consumer_offsets,0