Hi,
I keep seeing the following situation where a task is blocked getting a
MemorySegment from the pool but the TaskManager is reporting that it has
lots of MemorySegments available.
I'm completely stumped as to how to debug or what to look at next so any
hints/help/advice would be greatly
porary volumes in use are dedicated to the TaskManager and
> not shared :-)
> Yes, it is safe in your case.
>
> Best,
> Gary
>
> On Tue, Mar 10, 2020 at 6:39 PM David Maddison
> wrote:
>
>> Hi,
>>
>> When a TaskManager is restarted it can leave behind unr
Hi,
When a TaskManager is restarted it can leave behind unreferenced BlobServer
cache directories in the temporary storage that never get cleaned up.
Would it be safe to automatically clear the temporary storage every time
when a TaskManager is started?
(Note: the temporary volumes in use are
Hi KristoffSC,
As Zhu Zhu explained, Flink does not currently auto-scale a Job as new
resources become available. Instead the Job must be stopped via a savepoint
and restarted with a new parallelism (the old rescale CLI experiment use to
perform this).
Making Flink reactive to new resources and
the ttl and have a
> try.
> sun.net.inetaddr.ttl
> sun.net.inetaddr.negative.ttl
>
>
> Best,
> Yang
>
> David Maddison 于2019年11月29日周五 下午6:41写道：
>
>> I have a Flink 1.7 cluster using the "flink:1.7.2" (OpenJDK build
>> 1.8.0_222-b10) image on Kuberne
I have a Flink 1.7 cluster using the "flink:1.7.2" (OpenJDK build
1.8.0_222-b10) image on Kubernetes.
As part of a MasterRestoreHook (for checkpointing) the JobManager needs to
communicate with an external security service. This all works well until
there's a DNS lookup failure (due to network
