[
https://issues.apache.org/jira/browse/FLINK-15507?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17481139#comment-17481139
]
Piotr Nowojski edited comment on FLINK-15507 at 1/24/22, 2:36 PM:
------------------------------------------------------------------
I would like to resurrect the discussion about this ticket. As far as I was
told, enabling local recovery by default is still blocked by cleaning up some
files on {{notifyCheckpointAborted}} call implemented via FLINK-8871. Contrary
to what has been discussed in this ticket above
{quote}
A known issue when we enable local recovery by default is that local state
cannot be cleaned if no checkpoint ever completed, which is one of the goals
targeted by FLINK-8871.
{quote}
FLINK-8871 didn't provide such feature, only a possibility to implement such
one? However as far I as understand it, this conflicts with FLINK-24611, am I
right? Even if checkpoint 42 has been aborted, we should keep it files
available locally, since checkpoint 43 might re-use those files?
[~roman] [~yunta]?
was (Author: pnowojski):
I would like to resurrect the discussion about this ticket. As far as I was
told, enabling local recovery by default is still blocked by cleaning up some
files on {{notifyCheckpointAborted}} call implemented via FLINK-8871. Contrary
to what has been discussed in this ticket above
{quote}
A known issue when we enable local recovery by default is that local state
cannot be cleaned if no checkpoint ever completed, which is one of the goals
targeted by FLINK-8871.
{quote}
FLINK-8871 didn't provide such feature, only a possibility to implement such
one? However as far I as understand it, this conflicts with FLINK-24611, am I
right?
[~roman] [~yunta]?
> Activate local recovery for RocksDB backends by default
> -------------------------------------------------------
>
> Key: FLINK-15507
> URL: https://issues.apache.org/jira/browse/FLINK-15507
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / State Backends
> Reporter: Stephan Ewen
> Assignee: Yuan Mei
> Priority: Major
> Labels: auto-deprioritized-critical, auto-unassigned,
> pull-request-available
>
> For the RocksDB state backend, local recovery has no overhead when
> incremental checkpoints are used.
> It should be activated by default, because it greatly helps with recovery.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
