[
https://issues.apache.org/jira/browse/FLINK-12852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16879272#comment-16879272
]
Piotr Nowojski commented on FLINK-12852:
----------------------------------------
For option number #2, we would need a new RPC call from JobManager to
TaskManager, "task X is unblocked". Initial version could just send it once
"all tasks are running", which later can be refined to "once all downstream
tasks are running". Handling of "task X is unblocked" on the task managers
maybe is not that difficult, considering that we already have the buffers
rebalancing code. I think in the long term this would be the best solution if
we want to keep min/optimal buffers range.
[~StephanEwen] how is the yours number #4 different to what is proposed in the
current PR? It sounds like the current proposal is a superset of functionality
that you described. If buffer requirements are unable to be met, fail the task
with a timeout while we try to keep min required buffers vs optimal number of
buffers constraint. If that's annoyance for the user, he should adjust the
network config, without any changes in the behaviour (except of failure instead
of deadlock).
> Deadlock occurs when requiring exclusive buffer for RemoteInputChannel
> ----------------------------------------------------------------------
>
> Key: FLINK-12852
> URL: https://issues.apache.org/jira/browse/FLINK-12852
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Network
> Affects Versions: 1.7.2, 1.8.1, 1.9.0
> Reporter: Yun Gao
> Assignee: Yun Gao
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 1.9.0
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When running tests with an upstream vertex and downstream vertex, deadlock
> occurs when submitting the job:
> {code:java}
> "Sink: Unnamed (3/500)" #136 prio=5 os_prio=0 tid=0x00007f2cca81b000
> nid=0x38845 waiting on condition [0x00007f2cbe9fe000]
> java.lang.Thread.State: TIMED_WAITING (parking)
> at sun.misc.Unsafe.park(Native Method)
> - parking to wait for <0x000000073ed6b6f0> (a
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
> at java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:233)
> at
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.awaitNanos(AbstractQueuedSynchronizer.java:2078)
> at java.util.concurrent.ArrayBlockingQueue.poll(ArrayBlockingQueue.java:418)
> at
> org.apache.flink.runtime.io.network.buffer.NetworkBufferPool.requestMemorySegments(NetworkBufferPool.java:180)
> at
> org.apache.flink.runtime.io.network.buffer.NetworkBufferPool.requestMemorySegments(NetworkBufferPool.java:54)
> at
> org.apache.flink.runtime.io.network.partition.consumer.RemoteInputChannel.assignExclusiveSegments(RemoteInputChannel.java:139)
> at
> org.apache.flink.runtime.io.network.partition.consumer.SingleInputGate.assignExclusiveSegments(SingleInputGate.java:312)
> - locked <0x000000073fbc81f0> (a java.lang.Object)
> at
> org.apache.flink.runtime.io.network.partition.consumer.SingleInputGate.setup(SingleInputGate.java:220)
> at
> org.apache.flink.runtime.taskmanager.Task.setupPartionsAndGates(Task.java:836)
> at org.apache.flink.runtime.taskmanager.Task.run(Task.java:598)
> at java.lang.Thread.run(Thread.java:834)
> {code}
> This is due to the required and max of local buffer pool is not the same and
> there may be over-allocation, when assignExclusiveSegments there are no
> available memory.
>
> The detail of the scenarios is as follows: The parallelism of both upstream
> vertex and downstream vertex are 1500 and 500 respectively. There are 200 TM
> and each TM has 10696 buffers( in total and has 10 slots. For a TM that runs
> 9 upstream tasks and 1 downstream task, the 9 upstream tasks start first with
> local buffer pool \{required = 500, max = 2 * 500 + 8 = 1008}, it produces
> data quickly and each occupy about 990 buffers. Then the DownStream task
> starts and try to assigning exclusive buffers for 1500 -9 = 1491
> InputChannels. It requires 2981 buffers but only 1786 left. Since not all
> downstream tasks can start, the job will be blocked finally and no buffer can
> be released, and the deadlock finally occurred.
>
> I think although increasing the network memory solves the problem, the
> deadlock may not be acceptable. Fined grained resource management
> Flink-12761 can solve this problem, but AFAIK in 1.9 it will not include the
> network memory into the ResourceProfile.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
