[jira] [Commented] (FLINK-6762) Cannot rescale externalized incremental checkpoints

[ https://issues.apache.org/jira/browse/FLINK-6762?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16028449#comment-16028449 ] Stefan Richter commented on FLINK-6762: --- Officially, we currently do not want to support rescaling

[jira] [Commented] (FLINK-6761) Limitation for maximum state size per key in RocksDB backend

[ https://issues.apache.org/jira/browse/FLINK-6761?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16028363#comment-16028363 ] Stefan Richter commented on FLINK-6761: --- Opened a corresponding issue in the RocksDB tracker:

[jira] [Updated] (FLINK-6761) Limitation for maximum state size per key in RocksDB backend

[ https://issues.apache.org/jira/browse/FLINK-6761?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6761: -- Description: RocksDB`s JNI bridge allows for putting and getting {{byte[]}} as keys and

[jira] [Created] (FLINK-6761) Limitation for maximum state size per key in RocksDB backend

Stefan Richter created FLINK-6761: - Summary: Limitation for maximum state size per key in RocksDB backend Key: FLINK-6761 URL: https://issues.apache.org/jira/browse/FLINK-6761 Project: Flink

[jira] [Created] (FLINK-6761) Limitation for maximum state size per key in RocksDB backend

Stefan Richter created FLINK-6761: - Summary: Limitation for maximum state size per key in RocksDB backend Key: FLINK-6761 URL: https://issues.apache.org/jira/browse/FLINK-6761 Project: Flink

[jira] [Commented] (FLINK-6333) Utilize Bloomfilters in RocksDb

[ https://issues.apache.org/jira/browse/FLINK-6333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16028212#comment-16028212 ] Stefan Richter commented on FLINK-6333: --- This issue still exists after some testing and stops us

[jira] [Closed] (FLINK-5756) When there are many values under the same key in ListState, RocksDBStateBackend performances poor

[ https://issues.apache.org/jira/browse/FLINK-5756?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-5756. - Resolution: Fixed Fix Version/s: 1.2.1 1.3.0 fixed in e651df99f4 >

[jira] [Closed] (FLINK-6690) Rescaling broken

[ https://issues.apache.org/jira/browse/FLINK-6690?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6690. - Resolution: Fixed fixed in 36830adace > Rescaling broken > > >

[jira] [Closed] (FLINK-6685) SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier

[ https://issues.apache.org/jira/browse/FLINK-6685?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6685. - Resolution: Fixed fixed in 7639d49e03 > SafetyNetCloseableRegistry is closed prematurely in >

[jira] [Commented] (FLINK-6690) Rescaling broken

[ https://issues.apache.org/jira/browse/FLINK-6690?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16022602#comment-16022602 ] Stefan Richter commented on FLINK-6690: --- I guess the rescaling IT case will already catch this, it

[jira] [Commented] (FLINK-6690) Rescaling broken

[ https://issues.apache.org/jira/browse/FLINK-6690?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16021694#comment-16021694 ] Stefan Richter commented on FLINK-6690: --- I have found the cause in a bug that was introduced as part

[jira] [Commented] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16021248#comment-16021248 ] Stefan Richter commented on FLINK-6537: --- So this is not strictly related to incremental checkpoints,

[jira] [Commented] (FLINK-6685) SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier

[ https://issues.apache.org/jira/browse/FLINK-6685?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16021192#comment-16021192 ] Stefan Richter commented on FLINK-6685: --- For the release, we can fix this by reusing the registry

[jira] [Created] (FLINK-6685) SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier

Stefan Richter created FLINK-6685: - Summary: SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier Key: FLINK-6685 URL: https://issues.apache.org/jira/browse/FLINK-6685

[jira] [Created] (FLINK-6685) SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier

Stefan Richter created FLINK-6685: - Summary: SafetyNetCloseableRegistry is closed prematurely in Task::triggerCheckpointBarrier Key: FLINK-6685 URL: https://issues.apache.org/jira/browse/FLINK-6685

[jira] [Created] (FLINK-6684) Remove AsyncCheckpointRunnable from StreamTask

Stefan Richter created FLINK-6684: - Summary: Remove AsyncCheckpointRunnable from StreamTask Key: FLINK-6684 URL: https://issues.apache.org/jira/browse/FLINK-6684 Project: Flink Issue Type:

[jira] [Created] (FLINK-6684) Remove AsyncCheckpointRunnable from StreamTask

Stefan Richter created FLINK-6684: - Summary: Remove AsyncCheckpointRunnable from StreamTask Key: FLINK-6684 URL: https://issues.apache.org/jira/browse/FLINK-6684 Project: Flink Issue Type:

[jira] [Commented] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020941#comment-16020941 ] Stefan Richter commented on FLINK-6537: --- Ok, I think I have an idea about a possible problem, but it

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020789#comment-16020789 ] Stefan Richter edited comment on FLINK-6537 at 5/23/17 8:33 AM: I had a

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020789#comment-16020789 ] Stefan Richter edited comment on FLINK-6537 at 5/23/17 8:23 AM: I had a

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020789#comment-16020789 ] Stefan Richter edited comment on FLINK-6537 at 5/23/17 8:23 AM: I had a

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020789#comment-16020789 ] Stefan Richter edited comment on FLINK-6537 at 5/23/17 8:20 AM: I had a

[jira] [Commented] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16020789#comment-16020789 ] Stefan Richter commented on FLINK-6537: --- I had a look at those logs, but they seem strange to me.

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16019717#comment-16019717 ] Stefan Richter edited comment on FLINK-6537 at 5/22/17 4:19 PM: Which

[jira] [Comment Edited] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16019717#comment-16019717 ] Stefan Richter edited comment on FLINK-6537 at 5/22/17 3:42 PM: Which

[jira] [Commented] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16019717#comment-16019717 ] Stefan Richter commented on FLINK-6537: --- Which exception is that exactly? And how does it happen all

[jira] [Commented] (FLINK-6633) Register with shared state registry before adding to CompletedCheckpointStore

[ https://issues.apache.org/jira/browse/FLINK-6633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16019373#comment-16019373 ] Stefan Richter commented on FLINK-6633: --- This is fixed through the issue you mentioned. > Register

[jira] [Commented] (FLINK-6620) Add KeyGroupCheckpointedOperator interface that works for checkpointing key-groups

[ https://issues.apache.org/jira/browse/FLINK-6620?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16019298#comment-16019298 ] Stefan Richter commented on FLINK-6620: --- Yes, and I think there problem with designing the API still

[jira] [Closed] (FLINK-6651) Clearing registeredStates map should be protected in SharedStateRegistry#clear

[ https://issues.apache.org/jira/browse/FLINK-6651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6651. - Resolution: Fixed Fix Version/s: 1.3.0 fixed in 392bc71301 > Clearing registeredStates

[jira] [Assigned] (FLINK-6651) Clearing registeredStates map should be protected in SharedStateRegistry#clear

[ https://issues.apache.org/jira/browse/FLINK-6651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter reassigned FLINK-6651: - Assignee: Stefan Richter > Clearing registeredStates map should be protected in

[jira] [Closed] (FLINK-6640) Ensure registration of shared state happens before externalizing a checkpoint

[ https://issues.apache.org/jira/browse/FLINK-6640?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6640. - Resolution: Fixed fixed in 040356391f > Ensure registration of shared state happens before

[jira] [Created] (FLINK-6640) Ensure registration of shared state happens before externalizing a checkpoint

Stefan Richter created FLINK-6640: - Summary: Ensure registration of shared state happens before externalizing a checkpoint Key: FLINK-6640 URL: https://issues.apache.org/jira/browse/FLINK-6640

[jira] [Created] (FLINK-6640) Ensure registration of shared state happens before externalizing a checkpoint

Stefan Richter created FLINK-6640: - Summary: Ensure registration of shared state happens before externalizing a checkpoint Key: FLINK-6640 URL: https://issues.apache.org/jira/browse/FLINK-6640

[jira] [Closed] (FLINK-6633) Register with shared state registry before adding to CompletedCheckpointStore

[ https://issues.apache.org/jira/browse/FLINK-6633?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6633. - Resolution: Fixed fixed in 0162543ac1 > Register with shared state registry before adding to

[jira] [Updated] (FLINK-6633) Register with shared state registry before adding to CompletedCheckpointStore

[ https://issues.apache.org/jira/browse/FLINK-6633?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6633: -- Summary: Register with shared state registry before adding to CompletedCheckpointStore (was:

[jira] [Created] (FLINK-6633) Register with shared state registry before aadding to CompletedCheckpointStore

Stefan Richter created FLINK-6633: - Summary: Register with shared state registry before aadding to CompletedCheckpointStore Key: FLINK-6633 URL: https://issues.apache.org/jira/browse/FLINK-6633

[jira] [Created] (FLINK-6633) Register with shared state registry before aadding to CompletedCheckpointStore

Stefan Richter created FLINK-6633: - Summary: Register with shared state registry before aadding to CompletedCheckpointStore Key: FLINK-6633 URL: https://issues.apache.org/jira/browse/FLINK-6633

[jira] [Created] (FLINK-6599) Communicate last committed checkpoint through trigger checkpoint message

Stefan Richter created FLINK-6599: - Summary: Communicate last committed checkpoint through trigger checkpoint message Key: FLINK-6599 URL: https://issues.apache.org/jira/browse/FLINK-6599 Project:

[jira] [Created] (FLINK-6599) Communicate last committed checkpoint through trigger checkpoint message

Stefan Richter created FLINK-6599: - Summary: Communicate last committed checkpoint through trigger checkpoint message Key: FLINK-6599 URL: https://issues.apache.org/jira/browse/FLINK-6599 Project:

[jira] [Commented] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16011952#comment-16011952 ] Stefan Richter commented on FLINK-6537: --- Hi, thanks a lot for the testing effort and yes, there is

[jira] [Updated] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

[ https://issues.apache.org/jira/browse/FLINK-6537?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6537: -- Priority: Major (was: Blocker) > Umbrella issue for fixes to incremental snapshots >

[jira] [Updated] (FLINK-6505) Proactively cleanup local FS for RocksDBKeyedStateBackend on startup

[ https://issues.apache.org/jira/browse/FLINK-6505?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6505: -- Priority: Major (was: Blocker) > Proactively cleanup local FS for RocksDBKeyedStateBackend on

[jira] [Updated] (FLINK-6540) Add more thorough tests for RocksDB and incremental checkpointing

[ https://issues.apache.org/jira/browse/FLINK-6540?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6540: -- Priority: Major (was: Blocker) > Add more thorough tests for RocksDB and incremental

[jira] [Closed] (FLINK-6545) Make incremental checkpoints externalizable

[ https://issues.apache.org/jira/browse/FLINK-6545?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6545. - Resolution: Fixed fixed in 098e46f2d2 > Make incremental checkpoints externalizable >

[jira] [Closed] (FLINK-6535) JobID should not be part of the registration key to the SharedStateRegistry

[ https://issues.apache.org/jira/browse/FLINK-6535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6535. - Resolution: Fixed fixed in 4745d0c082 > JobID should not be part of the registration key to the

[jira] [Closed] (FLINK-6534) SharedStateRegistry is disposing state handles from main thread

[ https://issues.apache.org/jira/browse/FLINK-6534?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6534. - Resolution: Fixed fixed in 44fb035e02. > SharedStateRegistry is disposing state handles from

[jira] [Closed] (FLINK-6533) Duplicated registration of new shared state when checkpoint confirmations are still pending

[ https://issues.apache.org/jira/browse/FLINK-6533?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6533. - Resolution: Fixed fixed in 4745d0c082 > Duplicated registration of new shared state when

[jira] [Closed] (FLINK-6527) OperatorSubtaskState has empty implementations of (un)/registerSharedStates

[ https://issues.apache.org/jira/browse/FLINK-6527?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6527. - Resolution: Fixed fixed in efbb41bc63 > OperatorSubtaskState has empty implementations of

[jira] [Closed] (FLINK-6504) Lack of synchronization on materializedSstFiles in RocksDBKEyedStateBackend

[ https://issues.apache.org/jira/browse/FLINK-6504?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6504. - Resolution: Fixed fixed in 958773b71c > Lack of synchronization on materializedSstFiles in

[jira] [Commented] (FLINK-5053) Incremental / lightweight snapshots for checkpoints

[ https://issues.apache.org/jira/browse/FLINK-5053?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16006995#comment-16006995 ] Stefan Richter commented on FLINK-5053: --- I have opened this jira to track everything that should

[jira] [Created] (FLINK-6556) Include and expose Flink version in the savepoint data

Stefan Richter created FLINK-6556: - Summary: Include and expose Flink version in the savepoint data Key: FLINK-6556 URL: https://issues.apache.org/jira/browse/FLINK-6556 Project: Flink Issue

[jira] [Created] (FLINK-6556) Include and expose Flink version in the savepoint data

Stefan Richter created FLINK-6556: - Summary: Include and expose Flink version in the savepoint data Key: FLINK-6556 URL: https://issues.apache.org/jira/browse/FLINK-6556 Project: Flink Issue

[jira] [Created] (FLINK-6545) Make incremental checkpoints externalizable

Stefan Richter created FLINK-6545: - Summary: Make incremental checkpoints externalizable Key: FLINK-6545 URL: https://issues.apache.org/jira/browse/FLINK-6545 Project: Flink Issue Type:

[jira] [Created] (FLINK-6545) Make incremental checkpoints externalizable

Stefan Richter created FLINK-6545: - Summary: Make incremental checkpoints externalizable Key: FLINK-6545 URL: https://issues.apache.org/jira/browse/FLINK-6545 Project: Flink Issue Type:

[jira] [Created] (FLINK-6540) Create tests that cover all the issues

Stefan Richter created FLINK-6540: - Summary: Create tests that cover all the issues Key: FLINK-6540 URL: https://issues.apache.org/jira/browse/FLINK-6540 Project: Flink Issue Type: Sub-task

[jira] [Created] (FLINK-6540) Create tests that cover all the issues

Stefan Richter created FLINK-6540: - Summary: Create tests that cover all the issues Key: FLINK-6540 URL: https://issues.apache.org/jira/browse/FLINK-6540 Project: Flink Issue Type: Sub-task

[jira] [Updated] (FLINK-6475) Incremental snapshots in RocksDB hold lock during async file upload

[ https://issues.apache.org/jira/browse/FLINK-6475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6475: -- Issue Type: Sub-task (was: Bug) Parent: FLINK-6537 > Incremental snapshots in RocksDB

[jira] [Updated] (FLINK-6504) Lack of synchronization on materializedSstFiles in RocksDBKEyedStateBackend

[ https://issues.apache.org/jira/browse/FLINK-6504?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6504: -- Issue Type: Sub-task (was: Bug) Parent: FLINK-6537 > Lack of synchronization on

[jira] [Updated] (FLINK-6505) Proactively cleanup local FS for RocksDBKeyedStateBackend on startup

[ https://issues.apache.org/jira/browse/FLINK-6505?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6505: -- Issue Type: Sub-task (was: Bug) Parent: FLINK-6537 > Proactively cleanup local FS for

[jira] [Updated] (FLINK-6527) OperatorSubtaskState has empty implementations of (un)/registerSharedStates

[ https://issues.apache.org/jira/browse/FLINK-6527?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6527: -- Issue Type: Sub-task (was: Bug) Parent: FLINK-6537 > OperatorSubtaskState has empty

[jira] [Updated] (FLINK-6533) Duplicated registration of new shared state when checkpoint confirmations are still pending

[ https://issues.apache.org/jira/browse/FLINK-6533?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6533: -- Issue Type: Sub-task (was: Bug) Parent: FLINK-6537 > Duplicated registration of new

[jira] [Updated] (FLINK-6534) SharedStateRegistry is disposing state handles from main thread

[ https://issues.apache.org/jira/browse/FLINK-6534?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6534: -- Issue Type: Sub-task (was: Improvement) Parent: FLINK-6537 > SharedStateRegistry is

[jira] [Updated] (FLINK-6535) JobID should not be part of the registration key to the SharedStateRegistry

[ https://issues.apache.org/jira/browse/FLINK-6535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6535: -- Issue Type: Sub-task (was: Improvement) Parent: FLINK-6537 > JobID should not be part

[jira] [Created] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

Stefan Richter created FLINK-6537: - Summary: Umbrella issue for fixes to incremental snapshots Key: FLINK-6537 URL: https://issues.apache.org/jira/browse/FLINK-6537 Project: Flink Issue

[jira] [Created] (FLINK-6537) Umbrella issue for fixes to incremental snapshots

Stefan Richter created FLINK-6537: - Summary: Umbrella issue for fixes to incremental snapshots Key: FLINK-6537 URL: https://issues.apache.org/jira/browse/FLINK-6537 Project: Flink Issue

[jira] [Created] (FLINK-6535) JobID should not be part of the registration key to the SharedStateRegistry

Stefan Richter created FLINK-6535: - Summary: JobID should not be part of the registration key to the SharedStateRegistry Key: FLINK-6535 URL: https://issues.apache.org/jira/browse/FLINK-6535 Project:

[jira] [Created] (FLINK-6535) JobID should not be part of the registration key to the SharedStateRegistry

Stefan Richter created FLINK-6535: - Summary: JobID should not be part of the registration key to the SharedStateRegistry Key: FLINK-6535 URL: https://issues.apache.org/jira/browse/FLINK-6535 Project:

[jira] [Created] (FLINK-6534) SharedStateRegistry is disposing state handles from main thread

Stefan Richter created FLINK-6534: - Summary: SharedStateRegistry is disposing state handles from main thread Key: FLINK-6534 URL: https://issues.apache.org/jira/browse/FLINK-6534 Project: Flink

[jira] [Created] (FLINK-6534) SharedStateRegistry is disposing state handles from main thread

Stefan Richter created FLINK-6534: - Summary: SharedStateRegistry is disposing state handles from main thread Key: FLINK-6534 URL: https://issues.apache.org/jira/browse/FLINK-6534 Project: Flink

[jira] [Created] (FLINK-6533) Duplicated registration of new shared state when checkpoint confirmations are still pending

Stefan Richter created FLINK-6533: - Summary: Duplicated registration of new shared state when checkpoint confirmations are still pending Key: FLINK-6533 URL: https://issues.apache.org/jira/browse/FLINK-6533

[jira] [Created] (FLINK-6533) Duplicated registration of new shared state when checkpoint confirmations are still pending

Stefan Richter created FLINK-6533: - Summary: Duplicated registration of new shared state when checkpoint confirmations are still pending Key: FLINK-6533 URL: https://issues.apache.org/jira/browse/FLINK-6533

[jira] [Created] (FLINK-6527) OperatorSubtaskState has empty implementations of (un)/registerSharedStates

Stefan Richter created FLINK-6527: - Summary: OperatorSubtaskState has empty implementations of (un)/registerSharedStates Key: FLINK-6527 URL: https://issues.apache.org/jira/browse/FLINK-6527 Project:

[jira] [Created] (FLINK-6527) OperatorSubtaskState has empty implementations of (un)/registerSharedStates

Stefan Richter created FLINK-6527: - Summary: OperatorSubtaskState has empty implementations of (un)/registerSharedStates Key: FLINK-6527 URL: https://issues.apache.org/jira/browse/FLINK-6527 Project:

[jira] [Created] (FLINK-6507) Potential bug in ContinuousFileProcessing

Stefan Richter created FLINK-6507: - Summary: Potential bug in ContinuousFileProcessing Key: FLINK-6507 URL: https://issues.apache.org/jira/browse/FLINK-6507 Project: Flink Issue Type: Bug

[jira] [Created] (FLINK-6507) Potential bug in ContinuousFileProcessing

Stefan Richter created FLINK-6507: - Summary: Potential bug in ContinuousFileProcessing Key: FLINK-6507 URL: https://issues.apache.org/jira/browse/FLINK-6507 Project: Flink Issue Type: Bug

[jira] [Closed] (FLINK-6506) Some tests in flink-tests exceed the memory resources in containerized Travis builds

[ https://issues.apache.org/jira/browse/FLINK-6506?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6506. - Resolution: Fixed Fixed in 6b451d949a. > Some tests in flink-tests exceed the memory resources

[jira] [Updated] (FLINK-6506) Some tests in flink-tests exceed the memory resources in containerized Travis builds

[ https://issues.apache.org/jira/browse/FLINK-6506?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6506: -- Summary: Some tests in flink-tests exceed the memory resources in containerized Travis builds

[jira] [Created] (FLINK-6506) Some tests in flink-tests exceed memory the resources in containerized Travis builds

Stefan Richter created FLINK-6506: - Summary: Some tests in flink-tests exceed memory the resources in containerized Travis builds Key: FLINK-6506 URL: https://issues.apache.org/jira/browse/FLINK-6506

[jira] [Created] (FLINK-6506) Some tests in flink-tests exceed memory the resources in containerized Travis builds

Stefan Richter created FLINK-6506: - Summary: Some tests in flink-tests exceed memory the resources in containerized Travis builds Key: FLINK-6506 URL: https://issues.apache.org/jira/browse/FLINK-6506

[jira] [Updated] (FLINK-6505) Proactively cleanup local FS for RocksDBKeyedStateBackend on startup

[ https://issues.apache.org/jira/browse/FLINK-6505?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6505: -- Description: In {{RocksDBKeyedStateBackend}}, the {{instanceBasePath}} is cleared on

[jira] [Updated] (FLINK-6504) Lack of synchronization on materializedSstFiles in RocksDBKEyedStateBackend

[ https://issues.apache.org/jira/browse/FLINK-6504?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6504: -- Affects Version/s: 1.3.0 Component/s: State Backends, Checkpointing > Lack of

[jira] [Created] (FLINK-6505) Proactively cleanup local FS for RocksDBKeyedStateBackend on startup

Stefan Richter created FLINK-6505: - Summary: Proactively cleanup local FS for RocksDBKeyedStateBackend on startup Key: FLINK-6505 URL: https://issues.apache.org/jira/browse/FLINK-6505 Project: Flink

[jira] [Created] (FLINK-6505) Proactively cleanup local FS for RocksDBKeyedStateBackend on startup

Stefan Richter created FLINK-6505: - Summary: Proactively cleanup local FS for RocksDBKeyedStateBackend on startup Key: FLINK-6505 URL: https://issues.apache.org/jira/browse/FLINK-6505 Project: Flink

[jira] [Created] (FLINK-6504) Lack of synchronization on materializedSstFiles in RocksDBKEyedStateBackend

Stefan Richter created FLINK-6504: - Summary: Lack of synchronization on materializedSstFiles in RocksDBKEyedStateBackend Key: FLINK-6504 URL: https://issues.apache.org/jira/browse/FLINK-6504 Project:

[jira] [Updated] (FLINK-6484) Send only the registry keys for already registered files in incremental checkpointing

[ https://issues.apache.org/jira/browse/FLINK-6484?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter updated FLINK-6484: -- Description: State handles for files that are (potentially) shared across multiple incremental

[jira] [Created] (FLINK-6485) Use buffering to avoid frequent memtable flushes for short intervals in RockdDB incremental checkpoints

Stefan Richter created FLINK-6485: - Summary: Use buffering to avoid frequent memtable flushes for short intervals in RockdDB incremental checkpoints Key: FLINK-6485 URL:

[jira] [Created] (FLINK-6485) Use buffering to avoid frequent memtable flushes for short intervals in RockdDB incremental checkpoints

Stefan Richter created FLINK-6485: - Summary: Use buffering to avoid frequent memtable flushes for short intervals in RockdDB incremental checkpoints Key: FLINK-6485 URL:

[jira] [Created] (FLINK-6484) Send only the registry keys for already registered files in incremental checkpointing

Stefan Richter created FLINK-6484: - Summary: Send only the registry keys for already registered files in incremental checkpointing Key: FLINK-6484 URL: https://issues.apache.org/jira/browse/FLINK-6484

[jira] [Created] (FLINK-6484) Send only the registry keys for already registered files in incremental checkpointing

Stefan Richter created FLINK-6484: - Summary: Send only the registry keys for already registered files in incremental checkpointing Key: FLINK-6484 URL: https://issues.apache.org/jira/browse/FLINK-6484

[jira] [Created] (FLINK-6481) Encapsulate different snapshot algorithms (full, incremental) slightly clearer into strategy pattern

Stefan Richter created FLINK-6481: - Summary: Encapsulate different snapshot algorithms (full, incremental) slightly clearer into strategy pattern Key: FLINK-6481 URL:

[jira] [Created] (FLINK-6481) Encapsulate different snapshot algorithms (full, incremental) slightly clearer into strategy pattern

Stefan Richter created FLINK-6481: - Summary: Encapsulate different snapshot algorithms (full, incremental) slightly clearer into strategy pattern Key: FLINK-6481 URL:

[jira] [Created] (FLINK-6480) Add executor service parameter Snapshotable::snapshot(...)

Stefan Richter created FLINK-6480: - Summary: Add executor service parameter Snapshotable::snapshot(...) Key: FLINK-6480 URL: https://issues.apache.org/jira/browse/FLINK-6480 Project: Flink

[jira] [Created] (FLINK-6480) Add executor service parameter Snapshotable::snapshot(...)

Stefan Richter created FLINK-6480: - Summary: Add executor service parameter Snapshotable::snapshot(...) Key: FLINK-6480 URL: https://issues.apache.org/jira/browse/FLINK-6480 Project: Flink

[jira] [Closed] (FLINK-6471) RocksDBStateBackendTest#testCancelRunningSnapshot sometimes fails

[ https://issues.apache.org/jira/browse/FLINK-6471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-6471. - Resolution: Fixed Assignee: Stefan Richter Fix Version/s: 1.3.0 Fixed in

[jira] [Created] (FLINK-6475) Incremental snapshots in RocksDB hold lock during async file upload

Stefan Richter created FLINK-6475: - Summary: Incremental snapshots in RocksDB hold lock during async file upload Key: FLINK-6475 URL: https://issues.apache.org/jira/browse/FLINK-6475 Project: Flink

[jira] [Created] (FLINK-6475) Incremental snapshots in RocksDB hold lock during async file upload

Stefan Richter created FLINK-6475: - Summary: Incremental snapshots in RocksDB hold lock during async file upload Key: FLINK-6475 URL: https://issues.apache.org/jira/browse/FLINK-6475 Project: Flink

[jira] [Commented] (FLINK-6471) RocksDBStateBackendTest#testCancelRunningSnapshot sometimes fails

[ https://issues.apache.org/jira/browse/FLINK-6471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15999837#comment-15999837 ] Stefan Richter commented on FLINK-6471: --- Thanks for reporting this problem. This is cause by a

[jira] [Commented] (FLINK-5599) State interface docs often refer to keyed state only

[ https://issues.apache.org/jira/browse/FLINK-5599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15998370#comment-15998370 ] Stefan Richter commented on FLINK-5599: --- This looks like we can close it by now. The state docs have

[jira] [Closed] (FLINK-5599) State interface docs often refer to keyed state only

[ https://issues.apache.org/jira/browse/FLINK-5599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Stefan Richter closed FLINK-5599. - Resolution: Fixed > State interface docs often refer to keyed state only >

[jira] [Commented] (FLINK-5463) RocksDB.disposeInternal does not react to interrupts, blocks task cancellation

[ https://issues.apache.org/jira/browse/FLINK-5463?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15998368#comment-15998368 ] Stefan Richter commented on FLINK-5463: --- [~rmetzger] How should we proceed with this issue? >

<    4   5   6   7   8   9   10   11   12   >