[ 
https://issues.apache.org/jira/browse/FLINK-9114?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16428504#comment-16428504
 ] 

Aljoscha Krettek commented on FLINK-9114:
-----------------------------------------

This sounds like a good idea! Could you go a bit into details about where the 
{{CheckpointRecoveryFactory}} would fit in the the lifecycle of a checkpoint? 
I.e. who calls it and when, and how it will be created from a configuration in 
the factory.

> Enable user-provided, custom CheckpointRecoveryFactory for non-HA deployments
> -----------------------------------------------------------------------------
>
>                 Key: FLINK-9114
>                 URL: https://issues.apache.org/jira/browse/FLINK-9114
>             Project: Flink
>          Issue Type: Improvement
>          Components: Configuration, State Backends, Checkpointing
>            Reporter: Jacob Park
>            Assignee: Jacob Park
>            Priority: Major
>
> When you operate a Flink application that uses externalized checkpoints to 
> S3, it becomes difficult to determine which checkpoint is the latest to 
> recover from. Because S3 provides read-after-write consistency only for PUTS, 
> listing a S3 path is not guaranteed to be consistent, so we do not know what 
> checkpoint to recover from.
> The goal of this improvement is to allow users to provide a custom 
> CheckpointRecoveryFactory for non-HA deployments such that we can use this 
> feature to fail checkpoints if we cannot guarantee we will know where a 
> checkpoint will be in S3, and co-publish checkpoint metadata to a strongly 
> consistent data store.
> I propose the following changes:
>  # Modify AbstractNonHaServices and StandaloneHaServices to accept an 
> Executor for the custom CheckpointRecoveryFactory.
>  # Create a CheckpointRecoveryFactoryLoader to provide the custom 
> CheckpointRecoveryFactory from configurations.
>  # Add new configurations for this feature.
> We considered the pluggable StateBackend and potential pluggable 
> HighAvailabilityServices. These were too convoluted to solve our problem, so 
> we would like custom CheckpointRecoveryFactory.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to