[
https://issues.apache.org/jira/browse/FLINK-19832?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Till Rohrmann closed FLINK-19832.
---------------------------------
Resolution: Fixed
Fixed via
1.13.0:
58ed205390078929c0691196cf692029c837ea9c
f097b9387e99876f7b2a02a049a1b2783554390d
1.12.1:
b4155ecd2185b82cff713d4382f5245d661ec353
3e8448c65a50bc94676e6a916d9a6b18b12b7210
> Improve handling of immediately failed physical slot in
> SlotSharingExecutionSlotAllocator
> -----------------------------------------------------------------------------------------
>
> Key: FLINK-19832
> URL: https://issues.apache.org/jira/browse/FLINK-19832
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Affects Versions: 1.12.0
> Reporter: Andrey Zagrebin
> Assignee: Andrey Zagrebin
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.13.0, 1.12.1
>
>
> Improve handling of immediately failed physical slot in
> SlotSharingExecutionSlotAllocator
> If a physical slot future the immediately fails for a new SharedSlot in
> SlotSharingExecutionSlotAllocator#getOrAllocateSharedSlot but we continue to
> add logical slots to this SharedSlot, eventually, the logical slot also fails
> and gets removed from {{the SharedSlot}} which gets released (state
> RELEASED). The subsequent logical slot addings in the loop of
> {{allocateLogicalSlotsFromSharedSlots}} will fail the scheduling
> with the ALLOCATED state check because it will be RELEASED.
> The subsequent bulk timeout check will also not find the SharedSlot and fail
> with NPE.
> Hence, such SharedSlot with the immediately failed physical slot future
> should not be kept in the SlotSharingExecutionSlotAllocator and the logical
> slot requests depending on it can be immediately returned failed. The bulk
> timeout check does not need to be started because if some physical (and its
> logical) slot requests failed then the whole bulk will be canceled by
> scheduler.
> If the last assumption is not true for the future scheduling, this bulk
> failure might need additional explicit pending requests cancelation. We
> expect to refactor it for the declarative scheduling anyways.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
