[
https://issues.apache.org/jira/browse/FLINK-24611?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435647#comment-17435647
]
Roman Khachatryan commented on FLINK-24611:
-------------------------------------------
Thanks for the feedback!
I agree that option (1) (traversing the entries) is better (though I'm not sure
how async disposal would work without some concurrent data structure, which
would likely slow down the main thread; but that optmization can be considered
later).
Regarding job cancellation and avoiding state removal, I planned to check
newJobStatus.isGloballyTerminalState() in
CheckpointCoordinatorDeActivator.jobStatusChanges; if it's true then cleanup
pending checkpoints.
As for the removal of unused state after JM failover, I think it actually would
be a regression: currently, JM discards all the state on abort/suspend
immediately. Or do you mean, that state artifacts may be left in general
currently?
And after reading your comment, I see another issue: TMs will send
PlaceholderStreamStateHandle for checkpoint X+1, but the new JM won't have the
relevant entry.
To solve this, I see the following options:
# send normal handles from TMs after JM failover (TMs reconnect currently;
notifying state backends seems do-able but not very pretty)
# send normal handles from TMs until JM confirms the checkpoint (similar to
the current way, but without re-uploading)
# carry pending checkpoints from JM to JM along with completed ones
Option (2) seems easier and cleaner than (1);
Carrying pending checkpoints (3) additionally solves the previous problem of
removing unused state after JM failover. However, in the long run I think we
need a cleanup mechanism anyways.
WDYT?
> Prevent JM from discarding state on checkpoint abortion
> -------------------------------------------------------
>
> Key: FLINK-24611
> URL: https://issues.apache.org/jira/browse/FLINK-24611
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Checkpointing
> Affects Versions: 1.15.0
> Reporter: Roman Khachatryan
> Priority: Major
> Fix For: 1.15.0
>
>
> When a checkpoint is aborted, JM discards any state that was sent to it and
> wasn't used in other checkpoints. This forces incremental state backends to
> wait for confirmation from JM before re-using this state. For changelog
> backend this is even more critical.
> One approach proposed was to make backends/TMs responsible for their state,
> until it's not shared with other TMs, i.e. until rescaling (private/shared
> state ownership track: FLINK-23342 and more).
> However, that approach is quite invasive.
>
> An alternative solution would be:
> 1. SharedStateRegistry remembers the latest checkpoint for each shared state
> (instead of usage count currently)
> 2. CompletedCheckpointStore notifies it about the lowest valid checkpoint
> (on subsumption)
> 3. SharedStateRegistry then discards any state associated with the lower
> (subsumed/aborted) checkpoints
> So the aborted checkpoint can only be discarded after some subsequent
> successful checkpoint (which can mark state as used).
> Only JM code is changed.
>
> Implementation considerations.
> On subsumption, JM needs to find all the unused state and discard it.
> This can either be done by
> 1) simply traversing all entries; or by
> 2) maintaining a set of entries per checkpoint (e.g. SortedMap<Long,
> List<K>>). This allows to skip unnecessary traversal at the cost of higher
> memory usage
> In both cases:
> - each entry stores last checkpoint ID it was used in (long)
> - key is hashed (even with plain traversal, map.entrySet.iterator.remove()
> computes hash internally)
> Given the following constraints:
> - 10M state entries at most
> - 10 (retained) checkpoint at most
> - 10 checkpoints per second at most
> - state entry key takes 32B (usually UUID or two UUIDs)
> The extra space for (2) would be in order of 10M*32B=38Mb.
> The extra time for (1) would be in order of 10M * 10 checkpoints per second
> * ratio of outdated entries per checkpoint. Depending on the ratio and the
> hardware, this could take up to hundreds of ms per second, blocking the main
> thread.
> So approach (2) seems reasonable.
>
> The following cases shouldn't pose any difficulties:
> 1. Recovery, re-scaling, and state used by not all or by no tasks - we still
> register all states on recovery even after FLINK-22483/FLINK-24086
> 2. PlaceholderStreamStateHandles
> 3. Cross-task state sharing - not an issue as long as everything is managed
> by JM
> 4. Dependencies between SharedStateRegistry and CompletedCheckpointStore -
> simple after FLINK-24086
> 5. Multiple concurrent checkpoints (below)
> Consider the following case:
> (nr. concurrent checkpoints > 1)
> 1. checkpoint 1 starts, TM reports that it uses file f1; checkpoint 1 gets
> aborted - f1 is now tracked
> 2. savepoint 2 starts, it *will* use f1
> 3. checkpoint 3 starts and finishes; it does NOT use file f1
> When a checkpoint finishes, all pending checkpoints before it are aborted -
> but not savepoints.
> Savepoints currently are NOT incremental. And in the future, incremental
> savepoints shouldn't share any artifacts with checkpionts.
> The following should be kept in mind:
> 1. On job cancellation, state of aborted checkpoints should be cleaned up
> explicitly
> 2. Savepoints should be ignored and not change
> CheckpointStore.lowestCheckpointID
>
> For the end users, this change might render as a delay in discarding state of
> aborted checkpoints, which seems acceptable.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
