Kafka, mail # user - Graceful termination of kafka broker after draining all the data consumed

Graceful termination of kafka broker after draining all the data consumed

"Bae, Jae Hyeon" 2013-01-06, 07:35

Hi

If I want to terminate kafka broker gracefully. Before termination, it
should stop receiving the traffic from producers and wait until all
data will be consumed.

I don't think that kafka 0.7.x is supporting this feature. If I want
to implement this feature for myself, could you give me a brief sketch
of implementation?

Thank you
Best, Jae

Re: Graceful termination of kafka broker after draining all the data consumed

Jun Rao 2013-01-06, 16:52

In 0.7, one way to do this is to use a vip. All producers send data to the
vip. To decommission a broker, you first take the broker out of vip so no
new data will be produced to it. Then you let the consumer drain the data
(you can use ConsumerOffsetChecker to check if all data has been consumed).
Finally, you can shut down the broker.

This will be much easier in 0.8 because of replication.

Thanks,

Jun

On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]> wrote:

> Hi
>
> If I want to terminate kafka broker gracefully. Before termination, it
> should stop receiving the traffic from producers and wait until all
> data will be consumed.
>
> I don't think that kafka 0.7.x is supporting this feature. If I want
> to implement this feature for myself, could you give me a brief sketch
> of implementation?
>
> Thank you
> Best, Jae
>

Re: Graceful termination of kafka broker after draining all the data consumed

Neha Narkhede 2013-01-06, 18:36

In 0.8, we will provide a way for your to shutdown the broker in a
controlled fashion. What that would include is moving all the leaders away
from the broker so that it does not take any more produce requests. Once
that is done, you can shutdown the broker normally. You don't have to wait
until the consumers have pulled the data simply because they can consume
from the new leader.

Thanks,
Neha
On Sun, Jan 6, 2013 at 8:52 AM, Jun Rao <[EMAIL PROTECTED]> wrote:

> In 0.7, one way to do this is to use a vip. All producers send data to the
> vip. To decommission a broker, you first take the broker out of vip so no
> new data will be produced to it. Then you let the consumer drain the data
> (you can use ConsumerOffsetChecker to check if all data has been consumed).
> Finally, you can shut down the broker.
>
> This will be much easier in 0.8 because of replication.
>
> Thanks,
>
> Jun
>
> On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]>
> wrote:
>
> > Hi
> >
> > If I want to terminate kafka broker gracefully. Before termination, it
> > should stop receiving the traffic from producers and wait until all
> > data will be consumed.
> >
> > I don't think that kafka 0.7.x is supporting this feature. If I want
> > to implement this feature for myself, could you give me a brief sketch
> > of implementation?
> >
> > Thank you
> > Best, Jae
> >
>

Re: Graceful termination of kafka broker after draining all the data consumed

"Bae, Jae Hyeon" 2013-01-07, 19:18

0.8 sounds really great!

OK, I will try after you release stable build of 0.8

Thank you
Best, Jae

On Sun, Jan 6, 2013 at 10:36 AM, Neha Narkhede <[EMAIL PROTECTED]> wrote:
> In 0.8, we will provide a way for your to shutdown the broker in a
> controlled fashion. What that would include is moving all the leaders away
> from the broker so that it does not take any more produce requests. Once
> that is done, you can shutdown the broker normally. You don't have to wait
> until the consumers have pulled the data simply because they can consume
> from the new leader.
>
> Thanks,
> Neha
>
>
> On Sun, Jan 6, 2013 at 8:52 AM, Jun Rao <[EMAIL PROTECTED]> wrote:
>
>> In 0.7, one way to do this is to use a vip. All producers send data to the
>> vip. To decommission a broker, you first take the broker out of vip so no
>> new data will be produced to it. Then you let the consumer drain the data
>> (you can use ConsumerOffsetChecker to check if all data has been consumed).
>> Finally, you can shut down the broker.
>>
>> This will be much easier in 0.8 because of replication.
>>
>> Thanks,
>>
>> Jun
>>
>> On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]>
>> wrote:
>>
>> > Hi
>> >
>> > If I want to terminate kafka broker gracefully. Before termination, it
>> > should stop receiving the traffic from producers and wait until all
>> > data will be consumed.
>> >
>> > I don't think that kafka 0.7.x is supporting this feature. If I want
>> > to implement this feature for myself, could you give me a brief sketch
>> > of implementation?
>> >
>> > Thank you
>> > Best, Jae
>> >
>>

Re: Graceful termination of kafka broker after draining all the data consumed

王国栋 2013-02-17, 10:20

Hi Jun,

If we use high level producer based on zookeeper, how can we decommission a
broker without message loss?

Since we want to partition the log with IP, if all the brokers use the
same vip, we can not use the customized partition strategy.

Thanks.
On Mon, Jan 7, 2013 at 12:52 AM, Jun Rao <[EMAIL PROTECTED]> wrote:

> In 0.7, one way to do this is to use a vip. All producers send data to the
> vip. To decommission a broker, you first take the broker out of vip so no
> new data will be produced to it. Then you let the consumer drain the data
> (you can use ConsumerOffsetChecker to check if all data has been consumed).
> Finally, you can shut down the broker.
>
> This will be much easier in 0.8 because of replication.
>
> Thanks,
>
> Jun
>
> On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]>
> wrote:
>
> > Hi
> >
> > If I want to terminate kafka broker gracefully. Before termination, it
> > should stop receiving the traffic from producers and wait until all
> > data will be consumed.
> >
> > I don't think that kafka 0.7.x is supporting this feature. If I want
> > to implement this feature for myself, could you give me a brief sketch
> > of implementation?
> >
> > Thank you
> > Best, Jae
> >
>

--
Guodong Wang
王国栋

Re: Graceful termination of kafka broker after draining all the data consumed

Jun Rao 2013-02-18, 16:51

In 0.7, it's not very easy to decommission a broker using ZK based
producers. It's possible to do that with a vip (then you can't do
partitioning). In 0.8 (probably 0.8.1), you can use a tool to move all
partitions off a broker first and then decommission it.

Thanks,

Jun
On Sun, Feb 17, 2013 at 2:19 AM, Íõ¹ú¶° <[EMAIL PROTECTED]> wrote:

> Hi Jun,
>
> If we use high level producer based on zookeeper, how can we decommission a
> broker without message loss?
>
> Since we want to partition the log with IP, if all the brokers use the
> same vip, we can not use the customized partition strategy.
>
> Thanks.
>
>
> On Mon, Jan 7, 2013 at 12:52 AM, Jun Rao <[EMAIL PROTECTED]> wrote:
>
> > In 0.7, one way to do this is to use a vip. All producers send data to
> the
> > vip. To decommission a broker, you first take the broker out of vip so
> no
> > new data will be produced to it. Then you let the consumer drain the data
> > (you can use ConsumerOffsetChecker to check if all data has been
> consumed).
> > Finally, you can shut down the broker.
> >
> > This will be much easier in 0.8 because of replication.
> >
> > Thanks,
> >
> > Jun
> >
> > On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]>
> > wrote:
> >
> > > Hi
> > >
> > > If I want to terminate kafka broker gracefully. Before termination, it
> > > should stop receiving the traffic from producers and wait until all
> > > data will be consumed.
> > >
> > > I don't think that kafka 0.7.x is supporting this feature. If I want
> > > to implement this feature for myself, could you give me a brief sketch
> > > of implementation?
> > >
> > > Thank you
> > > Best, Jae
> > >
> >
>
>
>
> --
> Guodong Wang
> Íõ¹ú¶°
>

Re: Graceful termination of kafka broker after draining all the data consumed

王国栋 2013-02-19, 01:41

Thanks Jun. I will check more info about 0.8
On Tue, Feb 19, 2013 at 12:51 AM, Jun Rao <[EMAIL PROTECTED]> wrote:

> In 0.7, it's not very easy to decommission a broker using ZK based
> producers. It's possible to do that with a vip (then you can't do
> partitioning). In 0.8 (probably 0.8.1), you can use a tool to move all
> partitions off a broker first and then decommission it.
>
> Thanks,
>
> Jun
>
>
> On Sun, Feb 17, 2013 at 2:19 AM, 王国栋 <[EMAIL PROTECTED]> wrote:
>
> > Hi Jun,
> >
> > If we use high level producer based on zookeeper, how can we
> decommission a
> > broker without message loss?
> >
> > Since we want to partition the log with IP, if all the brokers use the
> > same vip, we can not use the customized partition strategy.
> >
> > Thanks.
> >
> >
> > On Mon, Jan 7, 2013 at 12:52 AM, Jun Rao <[EMAIL PROTECTED]> wrote:
> >
> > > In 0.7, one way to do this is to use a vip. All producers send data to
> > the
> > > vip. To decommission a broker, you first take the broker out of vip so
> > no
> > > new data will be produced to it. Then you let the consumer drain the
> data
> > > (you can use ConsumerOffsetChecker to check if all data has been
> > consumed).
> > > Finally, you can shut down the broker.
> > >
> > > This will be much easier in 0.8 because of replication.
> > >
> > > Thanks,
> > >
> > > Jun
> > >
> > > On Sat, Jan 5, 2013 at 11:34 PM, Bae, Jae Hyeon <[EMAIL PROTECTED]>
> > > wrote:
> > >
> > > > Hi
> > > >
> > > > If I want to terminate kafka broker gracefully. Before termination,
> it
> > > > should stop receiving the traffic from producers and wait until all
> > > > data will be consumed.
> > > >
> > > > I don't think that kafka 0.7.x is supporting this feature. If I want
> > > > to implement this feature for myself, could you give me a brief
> sketch
> > > > of implementation?
> > > >
> > > > Thank you
> > > > Best, Jae
> > > >
> > >
> >
> >
> >
> > --
> > Guodong Wang
> > 王国栋
> >
>

--
Guodong Wang
王国栋