Lin: Yes, I think so.
On Wed, Oct 14, 2015 at 1:48 PM, Lin Ma <[email protected]> wrote: > Thanks Dong. Then I bet it will take a while before its release. > > Lin > > > -----Original Message----- > From: Dong Lin [mailto:[email protected]] > Sent: Tuesday, October 13, 2015 11:21 PM > To: [email protected] > Subject: Re: Kafka Monitoring Framework > > Hey Lin, > > Thanks for your interest. As Kartik mentioned in the blog post, LinkedIn > Kafka team is developing a Kafka monitoring framework to help detect > problem that may only be found, for example, when you run server/client > with different versions for a long time using production traffic. We hope > it can be used in complement to existing unit test and Ducktape test to > make Kafka more robust. > > But we don't know when it can be released -- it is still in its early stage > of development. > > > Thanks, > Dong > > > > On Tue, Oct 13, 2015 at 10:47 PM, Lin Ma <[email protected]> wrote: > > > Hi, I read from website that effort has been put on developing Kafka > > monitoring framework recently (detailed below). > > Anyone knows when this framework can be released into kafka major > version? > > > > Thanks. > > > > > > > https://engineering.linkedin.com/apache-kafka/how-we_re-improving-and-advancing-kafka-linkedin > > Kafka Monitoring Framework > > We have recently started working on this effort to have a standardized > way > > to monitor our Kafka clusters. The idea here is to run a set of test > > applications which are producing and consuming data into kafka topics and > > validating the basic guarantees (order, guaranteed delivery, data > integrity > > etc.) in addition to the end to end latencies for publishing and > consuming > > data. This effort is independent of the monitoring that we already have > > where we monitor metrics emitted by the Kafka brokers. > > > > In addition to monitoring production systems, it will also be used for > > validating a new Kafka build that we might pick from the open source > trunk > > to deploy to our production clusters. This will also help us ensure that > > new versions of the kafka broker don't break existing older clients. > > >
