You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pulsar.apache.org by GitBox <gi...@apache.org> on 2018/11/08 11:14:27 UTC
[GitHub] cbornet opened a new issue #2962: Geo Replication and consumer
failover
cbornet opened a new issue #2962: Geo Replication and consumer failover
URL: https://github.com/apache/pulsar/issues/2962
Hi,
Could you give guidance on how to failover consumers in an async Geo Replication scenario ?
When a Pulsar cluster crashes in one region (or is shutdown for whatever reason, or is too slow), it's a common need to failover the producers and consumers to the replicated cluster. But AFAIK, the subscription offsets are not preserved during replication (same as Kafka) so there's a need to reset the subscription to the last received message. For Kafka, in Confluent Entreprise v5, the consumers will send their consumption timestamps to a dedicated topic that can then be used to get the offset on the replicated cluster. Is there something similar in Pulsar ? Or maybe something better ?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services