Glad to hear that!
Cheers,
Till
On Tue, Nov 17, 2020 at 5:35 AM Eleanore Jin wrote:
> Hi Till,
>
> Thanks for the response! The metrics I got from cadvisor and visualized
> via dashboard shipped by kubernetes. I actually run the flink job for the
> past 2 weeks and the memory usage has been sta
Hi Till,
Thanks for the response! The metrics I got from cadvisor and visualized via
dashboard shipped by kubernetes. I actually run the flink job for the past
2 weeks and the memory usage has been stabilized. There is no issue so far.
I still could not figure out the mystery why it was trending u
Hi Eleanore,
sorry for my late reply. The heap dump you have sent does not look
problematic. How do you measure the pod memory usage exactly? If you start
the Flink process with -Xms5120m -Xmx5120m then Flink should allocate 5120
MB of heap memory. Hence, this should be exactly what you are seeing
I also tried enable native memory tracking, via jcmd, here is the memory
breakdown: https://ibb.co/ssrZB4F
since job manager memory configuration for flink 1.10.2 only has
jobmanager.heap.size, and it only translates to heap settings, should I
also set -XX:MaxDirectMemorySize and -XX:MaxMetaspaceS
Hi Till,
please see the screenshot of heap dump: https://ibb.co/92Hzrpr
Thanks!
Eleanore
On Fri, Oct 23, 2020 at 9:25 AM Eleanore Jin wrote:
> Hi Till,
> Thanks a lot for the prompt response, please see below information.
>
> 1. how much memory assign to JM pod?
> 6g for container memory limit
Hi Till,
Thanks a lot for the prompt response, please see below information.
1. how much memory assign to JM pod?
6g for container memory limit, 5g for jobmanager.heap.size, I think this is
the only available jm memory configuration for flink 1.10.2
2. Have you tried with newer Flink versions?
I
Hi Eleanore,
how much memory did you assign to the JM pod? Maybe the limit is so high
that it takes a bit of time until GC is triggered. Have you tried whether
the same problem also occurs with newer Flink versions?
The difference between checkpoints enabled and disabled is that the JM
needs to d
Hi all,
I have a flink job running version 1.10.2, it simply read from a kafka
topic with 96 partitions, and output to another kafka topic.
It is running in k8s, with 1 JM (not in HA mode), 12 task managers each has
4 slots.
The checkpoint persists the snapshot to azure blob storage, checkpoints
