[jira] [Updated] (FLINK-36020) exception: Time should be non negative when enable buffer debloating mechanism

["siyuan zheng (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-36020?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

siyuan zheng updated FLINK-36020:
---------------------------------
    Summary: exception: Time should be non negative when enable buffer debloating mechanism  (was: exception: Time should be non negative)

> exception: Time should be non negative when enable buffer debloating mechanism
> ------------------------------------------------------------------------------
>
>                 Key: FLINK-36020
>                 URL: https://issues.apache.org/jira/browse/FLINK-36020
>             Project: Flink
>          Issue Type: Bug
>          Components: Runtime / Network
>    Affects Versions: 1.19.1
>            Reporter: siyuan zheng
>            Priority: Major
>
> Running a Flink job on YARN, setting 'taskmanager.network.memory.buffer-debloat.period' to true may cause this error to occur intermittently. However, increasing the value of 'taskmanager.network.memory.buffer-debloat.period' reduces the frequency of these errors.
> It appears that this issue may be caused by host clock synchronization. However, I believe that a rollback of the host time should not result in an error. Fault tolerance should be enhanced to handle this situation.
>  
> {code:java}
> java.lang.IllegalArgumentException: Time should be non negative    
>   at org.apache.flink.util.Preconditions.checkArgument(Preconditions.java:138)   
>   at org.apache.flink.runtime.throughput.ThroughputCalculator.calculateThroughput(ThroughputCalculator.java:80)    
>   at org.apache.flink.runtime.throughput.ThroughputCalculator.calculateThroughput(ThroughputCalculator.java:71)    
>   at org.apache.flink.runtime.io.network.partition.consumer.SingleInputGate.triggerDebloating(SingleInputGate.java:464)    
>   at org.apache.flink.runtime.taskmanager.InputGateWithMetrics.triggerDebloating(InputGateWithMetrics.java:90)    
>   at org.apache.flink.streaming.runtime.tasks.StreamTask.debloat(StreamTask.java:894)    
>   at org.apache.flink.streaming.runtime.tasks.StreamTask.lambda$null$6(StreamTask.java:885)    
>   at org.apache.flink.streaming.runtime.tasks.StreamTaskActionExecutor$1.runThrowing(StreamTaskActionExecutor.java:50)    
>   at org.apache.flink.streaming.runtime.tasks.mailbox.Mail.run(Mail.java:90)    
>   at org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.runMail(MailboxProcessor.java:398)    
>   at org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.processMailsWhenDefaultActionUnavailable(MailboxProcessor.java:367)    
>   at org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.processMail(MailboxProcessor.java:352)    
>   at org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.runMailboxLoop(MailboxProcessor.java:229)    
>   at org.apache.flink.streaming.runtime.tasks.StreamTask.runMailboxLoop(StreamTask.java:909)    
>   at org.apache.flink.streaming.runtime.tasks.StreamTask.invoke(StreamTask.java:858)    
>   at org.apache.flink.runtime.taskmanager.Task.runWithSystemExitMonitoring(Task.java:958)    
>   at org.apache.flink.runtime.taskmanager.Task.restoreAndInvoke(Task.java:937)    
>   at org.apache.flink.runtime.taskmanager.Task.doRun(Task.java:751)    
>   at org.apache.flink.runtime.taskmanager.Task.run(Task.java:566)    
>   at java.lang.Thread.run(Thread.java:745)
> {code}
>  
>  



--
This message was sent by Atlassian Jira
(v8.20.10#820010)