I was thinking about the situation where you have less brokers in the ISR list than the number set in min.insync.replicas.
My idea was that if I, as an administrator, for a given topic, want to favor durability over availability, then if that topic has less ISR than the value set in min.insync.replicas I may want to stop producing to the topic. In the way min.insync.replicas and ack work, I need to coordinate with all producers in order to achieve this. There is no way (or I don't know it) to globally enforce stop producing to a topic if it is under replicated. I don't see why, for the same topic, some producers might want get an error when the number of ISR is below min.insync.replicas while other producers don't. I think it could be more useful to be able to set that ALL producers should get an error when a given topic is under replicated so they stop producing, than for a single producer to get an error when ANY topic is under replicated. I don't have a lot of experience with Kafka so I may be missing some use cases. But I understand your point, min.insync.replicas setting should be understood as "if a producer wants to get an error when topics are under replicated, then how many replicas are enough for not raising an error?" On Thu, Jan 26, 2017 at 4:16 PM, Ewen Cheslack-Postava <e...@confluent.io> wrote: > The acks setting for the producer doesn't affect the final durability > guarantees. These are still enforced by the replication and min ISR > settings. Instead, the ack setting just lets the producer control how > durable the write is before *that producer* can consider the write > "complete", i.e. before it gets an ack. > > -Ewen > > On Tue, Jan 24, 2017 at 12:46 PM, Luciano Afranllie < > listas.luaf...@gmail.com> wrote: > > > Hi everybody > > > > I am trying to understand why Kafka let each individual producer, on a > > connection per connection basis, choose the tradeoff between availability > > and durability, honoring min.insync.replicas value only if producer uses > > ack=all. > > > > I mean, for a single topic, cluster administrators can't enforce messages > > to be stores in a minimum number of replicas without coordinating with > all > > producers to that topic so all of them use ack=all. > > > > Is there something that I am missing? Is there any other strategy to > > overcome this situation? > > > > Regards > > Luciano > > >
