[jira] [Commented] (FLINK-8531) Support separation of "Exclusive", "Shared" and "Task owned" state

[ https://issues.apache.org/jira/browse/FLINK-8531?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17320035#comment-17320035 ] fanrui commented on FLINK-8531: --- [~sewen] For some high-parallel Flink tasks, when hdfs Namenode is under

[jira] [Commented] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17305647#comment-17305647 ] fanrui commented on FLINK-21726: [~yunta] Thanks for your work on the frocksdb project. Most of our

[jira] [Commented] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17299490#comment-17299490 ] fanrui commented on FLINK-21726: [frocksdb pr|https://github.com/ververica/frocksdb/pull/25] > Fix

[jira] [Commented] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17299482#comment-17299482 ] fanrui commented on FLINK-21726: Hi [~trohrmann] , thanks for your comments. There is a probability of

[jira] [Comment Edited] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17299474#comment-17299474 ] fanrui edited comment on FLINK-21726 at 3/11/21, 10:28 AM: --- Hi [~gary.wu]

[jira] [Commented] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17299474#comment-17299474 ] fanrui commented on FLINK-21726: Hi [~gary.wu] Checkpoint stuck may be due to other reasons. If this

[jira] [Updated] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-21726: --- Description: h1. 1. Bug description: When RocksDB Checkpoint, it may be stuck in

[jira] [Updated] (FLINK-21726) Fix checkpoint stuck

[ https://issues.apache.org/jira/browse/FLINK-21726?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-21726: --- Component/s: Runtime / Checkpointing > Fix checkpoint stuck > > > Key:

[jira] [Created] (FLINK-21726) Fix checkpoint stuck

fanrui created FLINK-21726: -- Summary: Fix checkpoint stuck Key: FLINK-21726 URL: https://issues.apache.org/jira/browse/FLINK-21726 Project: Flink Issue Type: Bug Components: Runtime /

[jira] [Commented] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17299267#comment-17299267 ] fanrui commented on FLINK-21436: [~yunta] Thanks for your comments.  I have read FLINK-18203 and the

[jira] [Commented] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17294349#comment-17294349 ] fanrui commented on FLINK-21436: h1. Give some test results of increasing the

[jira] [Updated] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-21436: --- Attachment: JM 启动火焰图.svg > Speed ​​up the restore of UnionListState >

[jira] [Updated] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-21436: --- Attachment: akka timeout Exception.png > Speed ​​up the restore of UnionListState >

[jira] [Comment Edited] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17288821#comment-17288821 ] fanrui edited comment on FLINK-21436 at 2/23/21, 3:54 AM: -- [~yunta] Thanks for

[jira] [Commented] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17288821#comment-17288821 ] fanrui commented on FLINK-21436: [~yunta] Thanks for your comments. I tried to increase the value of

[jira] [Updated] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-21436: --- Description: h1. 1. Problem introduction and cause analysis Problem description: The duration of

[jira] [Commented] (FLINK-21436) Speed ​​up the restore of UnionListState

[ https://issues.apache.org/jira/browse/FLINK-21436?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17288319#comment-17288319 ] fanrui commented on FLINK-21436: After online verification: The parallelism of Source is set to 2000,

[jira] [Created] (FLINK-21436) Speed ​​up the restore of UnionListState

fanrui created FLINK-21436: -- Summary: Speed ​​up the restore of UnionListState Key: FLINK-21436 URL: https://issues.apache.org/jira/browse/FLINK-21436 Project: Flink Issue Type: Improvement

[jira] [Commented] (FLINK-16444) Count the read/write/seek/next latency of RocksDB as metrics

[ https://issues.apache.org/jira/browse/FLINK-16444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17265671#comment-17265671 ] fanrui commented on FLINK-16444: [~yunta] I think sampling is more reasonable. The current function is

[jira] [Commented] (FLINK-16444) Count the read/write/seek/next latency of RocksDB as metrics

[ https://issues.apache.org/jira/browse/FLINK-16444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17264751#comment-17264751 ] fanrui commented on FLINK-16444: +1, some latency metrics are very useful. Use flink-benckmark to make

[jira] [Updated] (FLINK-20911) Support configuration of RocksDB log level

[ https://issues.apache.org/jira/browse/FLINK-20911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-20911: --- Summary: Support configuration of RocksDB log level (was: Support to modify RocksDB LOG level) > Support

[jira] [Commented] (FLINK-20911) Support to modify RocksDB LOG level

[ https://issues.apache.org/jira/browse/FLINK-20911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17262467#comment-17262467 ] fanrui commented on FLINK-20911: [~yunta], thanks for your commets. I like to do it. Can you assigned

[jira] [Created] (FLINK-20912) Increase Log and Metric: Time consumed by Checkpoint Restore

fanrui created FLINK-20912: -- Summary: Increase Log and Metric: Time consumed by Checkpoint Restore Key: FLINK-20912 URL: https://issues.apache.org/jira/browse/FLINK-20912 Project: Flink Issue

[jira] [Created] (FLINK-20911) Support to modify RocksDB LOG level

fanrui created FLINK-20911: -- Summary: Support to modify RocksDB LOG level Key: FLINK-20911 URL: https://issues.apache.org/jira/browse/FLINK-20911 Project: Flink Issue Type: Improvement

[jira] [Commented] (FLINK-20496) RocksDB partitioned index filter option

[ https://issues.apache.org/jira/browse/FLINK-20496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17262125#comment-17262125 ] fanrui commented on FLINK-20496: This feature looks very good. Our production environment encountered

[jira] [Commented] (FLINK-19911) add read buffer for input stream

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17247359#comment-17247359 ] fanrui commented on FLINK-19911: [~yunta] Thanks for your comment. I wrote a

[jira] [Updated] (FLINK-19911) add read buffer for input stream

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-19911: --- Summary: add read buffer for input stream (was: Read checkpoint stream with buffer to speedup restore) >

[jira] [Commented] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17226512#comment-17226512 ] fanrui commented on FLINK-19911: [~yunta], thx. Could you have time to help review it? Thank you. >

[jira] [Comment Edited] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17226458#comment-17226458 ] fanrui edited comment on FLINK-19911 at 11/5/20, 2:22 AM: -- Hi, [~yunta] , Can

[jira] [Commented] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17226458#comment-17226458 ] fanrui commented on FLINK-19911: Hi, [~tangyun] , Can you assigned to me? I have submitted a PR. Thanks.

[jira] [Issue Comment Deleted] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-19911: --- Comment: was deleted (was: Flink job [code link|

[jira] [Commented] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17224458#comment-17224458 ] fanrui commented on FLINK-19911: Adjust the buffer size and do some more detailed tests on Flink job.

[jira] [Commented] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17224406#comment-17224406 ] fanrui commented on FLINK-19911: Flink job [code link|

[jira] [Updated] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-19911: --- Description: Heap StateBackend needs to serialize each Java Object into the file system during snapshot.

[jira] [Updated] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

[ https://issues.apache.org/jira/browse/FLINK-19911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-19911: --- Component/s: FileSystems > Read checkpoint stream with buffer to speedup restore >

[jira] [Created] (FLINK-19911) Read checkpoint stream with buffer to speedup restore

fanrui created FLINK-19911: -- Summary: Read checkpoint stream with buffer to speedup restore Key: FLINK-19911 URL: https://issues.apache.org/jira/browse/FLINK-19911 Project: Flink Issue Type:

[jira] [Commented] (FLINK-18503) bug occurs when `HeapReducingState#add` method handles null

[ https://issues.apache.org/jira/browse/FLINK-18503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17161938#comment-17161938 ] fanrui commented on FLINK-18503: Thanks for [~yunta] 's reply. I have read this issue. > bug occurs

[jira] [Created] (FLINK-18503) bug occurs when `HeapReducingState#add` method handles null

fanrui created FLINK-18503: -- Summary: bug occurs when `HeapReducingState#add` method handles null Key: FLINK-18503 URL: https://issues.apache.org/jira/browse/FLINK-18503 Project: Flink Issue Type:

[jira] [Commented] (FLINK-18473) Optimize RocksDB disk load balancing strategy

[ https://issues.apache.org/jira/browse/FLINK-18473?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17150680#comment-17150680 ] fanrui commented on FLINK-18473: In the case of large samples, random strategies can also ensure load

[jira] [Created] (FLINK-18473) Optimize RocksDB disk load balancing strategy

fanrui created FLINK-18473: -- Summary: Optimize RocksDB disk load balancing strategy Key: FLINK-18473 URL: https://issues.apache.org/jira/browse/FLINK-18473 Project: Flink Issue Type: Improvement

[jira] [Commented] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16925538#comment-16925538 ] fanrui commented on FLINK-13677: Hi, [~jark] ,can you review this pr? thanks > Translate "Monitoring

[jira] [Issue Comment Deleted] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13677?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] fanrui updated FLINK-13677: --- Comment: was deleted (was: [https://flink.apache.org/2019/06/05/flink-network-stack.html]

[jira] [Commented] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16921887#comment-16921887 ] fanrui commented on FLINK-13677: Hi,[~jark], Can you assigned to me? I will completed in today. Thanks.  

[jira] [Commented] (FLINK-13644) Translate "State Backends" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16906874#comment-16906874 ] fanrui commented on FLINK-13644:  > Translate "State Backends" page into Chinese >

[jira] [Commented] (FLINK-13644) Translate "State Backends" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16906125#comment-16906125 ] fanrui commented on FLINK-13644: Hi, The 

[jira] [Commented] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16904436#comment-16904436 ] fanrui commented on FLINK-13677: [https://flink.apache.org/2019/06/05/flink-network-stack.html]

[jira] [Commented] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16904111#comment-16904111 ] fanrui commented on FLINK-13677: Hi, May I translate this? > Translate "Monitoring Back Pressure" page

[jira] [Created] (FLINK-13677) Translate "Monitoring Back Pressure" page into Chinese

fanrui created FLINK-13677: -- Summary: Translate "Monitoring Back Pressure" page into Chinese Key: FLINK-13677 URL: https://issues.apache.org/jira/browse/FLINK-13677 Project: Flink Issue Type:

[jira] [Comment Edited] (FLINK-13644) Translate "State Backends" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16902609#comment-16902609 ] fanrui edited comment on FLINK-13644 at 8/8/19 2:56 AM: Hi,[~jark], Can you

[jira] [Commented] (FLINK-13644) Translate "State Backends" page into Chinese

[ https://issues.apache.org/jira/browse/FLINK-13644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16902609#comment-16902609 ] fanrui commented on FLINK-13644: Hi,[~jark], Can you  assigned to me? I will completed in next week, >

[jira] [Created] (FLINK-13644) Translate "State Backends" page into Chinese

fanrui created FLINK-13644: -- Summary: Translate "State Backends" page into Chinese Key: FLINK-13644 URL: https://issues.apache.org/jira/browse/FLINK-13644 Project: Flink Issue Type: Sub-task

<    1   2