[ 
https://issues.apache.org/jira/browse/FLINK-8487?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16390207#comment-16390207
 ] 

ASF GitHub Bot commented on FLINK-8487:
---------------------------------------

Github user StephanEwen commented on a diff in the pull request:

    https://github.com/apache/flink/pull/5656#discussion_r172979459
  
    --- Diff: 
flink-runtime/src/main/java/org/apache/flink/runtime/concurrent/FutureUtils.java
 ---
    @@ -223,6 +224,81 @@
                }
        }
     
    +   /**
    +    * Retry the given operation with the given delay in between successful 
completions where the
    +    * result does not match a given predicate.
    +    *
    +    * @param operation to retry
    +    * @param retryDelay delay between retries
    +    * @param deadline A deadline that specifies at what point we should 
stop retrying
    +    * @param acceptancePredicate Predicate to test whether the result is 
acceptable
    +    * @param scheduledExecutor executor to be used for the retry operation
    +    * @param <T> type of the result
    +    * @return Future which retries the given operation a given amount of 
times and delays the retry
    +    *   in case the predicate isn't matched
    +    */
    +   public static <T> CompletableFuture<T> retrySuccesfulWithDelay(
    +           final Supplier<CompletableFuture<T>> operation,
    +           final Time retryDelay,
    --- End diff --
    
    Deadline uses `Duration`, this method uses `Time`.


> State loss after multiple restart attempts
> ------------------------------------------
>
>                 Key: FLINK-8487
>                 URL: https://issues.apache.org/jira/browse/FLINK-8487
>             Project: Flink
>          Issue Type: Bug
>          Components: State Backends, Checkpointing
>    Affects Versions: 1.3.2
>            Reporter: Fabian Hueske
>            Priority: Blocker
>             Fix For: 1.5.0, 1.4.3
>
>
> A user [reported this 
> issue|https://lists.apache.org/thread.html/9dc9b719cf8449067ad01114fedb75d1beac7b4dff171acdcc24903d@%3Cuser.flink.apache.org%3E]
>  on the user@f.a.o mailing list and analyzed the situation.
> Scenario:
> - A program that reads from Kafka and computes counts in a keyed 15 minute 
> tumbling window.  StateBackend is RocksDB and checkpointing is enabled.
> {code}
> keyBy(0)
>         .timeWindow(Time.of(window_size, TimeUnit.MINUTES))
>         .allowedLateness(Time.of(late_by, TimeUnit.SECONDS))
>         .reduce(new ReduceFunction(), new WindowFunction())
> {code}
> - At some point HDFS went into a safe mode due to NameNode issues
> - The following exception was thrown
> {code}
> org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.ipc.StandbyException):
>  Operation category WRITE is not supported in state standby. Visit 
> https://s.apache.org/sbnn-error
>     ..................
>     at 
> org.apache.flink.runtime.fs.hdfs.HadoopFileSystem.mkdirs(HadoopFileSystem.java:453)
>         at 
> org.apache.flink.core.fs.SafetyNetWrapperFileSystem.mkdirs(SafetyNetWrapperFileSystem.java:111)
>         at 
> org.apache.flink.runtime.state.filesystem.FsCheckpointStreamFactory.createBasePath(FsCheckpointStreamFactory.java:132)
> {code}
> - The pipeline came back after a few restarts and checkpoint failures, after 
> the HDFS issues were resolved.
> - It was evident that operator state was lost. Either it was the Kafka 
> consumer that kept on advancing it's offset between a start and the next 
> checkpoint failure (a minute's worth) or the the operator that had partial 
> aggregates was lost. 
> The user did some in-depth analysis (see [mail 
> thread|https://lists.apache.org/thread.html/9dc9b719cf8449067ad01114fedb75d1beac7b4dff171acdcc24903d@%3Cuser.flink.apache.org%3E])
>  and might have (according to [~aljoscha]) identified the problem.
> [~stefanrichte...@gmail.com], can you have a look at this issue and check if 
> it is relevant?



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to