[
https://issues.apache.org/jira/browse/FLINK-24611?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17441281#comment-17441281
]
Stephan Ewen commented on FLINK-24611:
--------------------------------------
Good points that you brought up.
*Regarding Left-over state on JM failover*
My assumption was around hard losses of the JM. In those cases, there wouldn't
be any cleanup run on the JM, so the state would be orphaned and left over.
In summary, we have those cases:
* +Task failure or checkpoint cancelled:+ slight but not meaningful regression,
because state for pending checkpoint gets cleaned up with next checkpoint.
* +Hard JM Failure:+ No cleanup of state that was help by JM for pending
checkpoint. State was and still is orphaned and left over.
* +Graceful JM leader loss:* State in pending checkpoint was previously deleted
by JM when CheckpointCoordinator got suspended, but is now orphaned. =>
+Regression+
My feeling is that this is an OK regression for now, because we anyways want to
fix the case of orphaned states properly with a cleanup/scavenger process.
I can create a separate issue suggestion on that.
*Regarding Placeholder Handles*
I think Roman's option (2) is good (only sending placeholders when the original
handle is in an ack-ed checkpoint).
I was thinking if we can even get rid of the Placeholder Handles. How often are
they really used and ho helpful are they?
* Aren't they mainly for byte[] handles? I guess for handles referencing files,
the saved space in the RPC message is minimal.
* How much small incremental state do we have? Incremental state are currently
only SST files, how often are those below 20KBytes? And isn't the expectation
that we avoid these small files in the future?
So, in conclusion, maybe we can avoid this altogether? Maybe [~yunta] has some
insights about whether I am overlooking some case here.
> 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
(v8.20.1#820001)
