Hi Thomas,
The current calculation for network buffer size is
"Math.min(taskmanager.network.memory.max,
Math.max(taskmanager.network.memory.min, fraction * totalMem))".
Based on your below configuration, the result is just 32768 bytes (8 buffers)
from taskmanager.network.memory.min.
If you want
Truong Duc Kien created FLINK-10857:
---
Summary: Conflict between JMX and Prometheus Metrics reporter
Key: FLINK-10857
URL: https://issues.apache.org/jira/browse/FLINK-10857
Project: Flink
Is
Hi Piotr
I have extracted the API portion of the design and the google doc is here.
Please review and provide your feedback.
Thanks,
Xuefu
--
Sender:Xuefu
Sent at:2018 Nov 12 (Mon) 12:43
Recipient:Piotr Nowojski ; dev
Cc:Bowen
Hi,
I'm trying to understand the intention behind the size parameters for
network buffers, specifically max, min and fraction. The reason we are
looking at it is an attempt to make the memory allocation elastic, so that
memory is allocated according to the actual number of buffers required
(within
Hi,
yes, that is correct. The failure mapper is there to cause a failover event for
which we can then check i) that exactly-once or at-least-once is not violated,
depending on the expected semantics and ii) that the restore works at all ;-).
You might be able to reuse org.apache.flink.streaming
Sounds good. Perhaps it would also be good to allow the user to specify an
options factory in flink-conf.yaml for more flexibility?
Thomas
On Mon, Nov 12, 2018 at 9:48 AM Stefan Richter
wrote:
> Hi,
>
> Ufuk is right, for historical reasons there is currently only the
> programatic way but I th
Hi,
Ufuk is right, for historical reasons there is currently only the programatic
way but I think nothing speaks fundamentally against offering configuration via
config in the future (maybe just a lot of config keys must be introduced to
cover all options).
Best,
Stefan
> On 9. Nov 2018, at 2
Till Rohrmann created FLINK-10855:
-
Summary: CheckpointCoordinator does not delete checkpoint
directory of late/failed checkpoints
Key: FLINK-10855
URL: https://issues.apache.org/jira/browse/FLINK-10855
Till Rohrmann created FLINK-10856:
-
Summary: Harden resume from externalized checkpoint E2E test
Key: FLINK-10856
URL: https://issues.apache.org/jira/browse/FLINK-10856
Project: Flink
Issue T
vinoyang created FLINK-10853:
Summary: Provide end-to-end test for Kafka 0.9 connector
Key: FLINK-10853
URL: https://issues.apache.org/jira/browse/FLINK-10853
Project: Flink
Issue Type: Test
vinoyang created FLINK-10854:
Summary: Provide end-to-end test for Kafka 0.8 connector
Key: FLINK-10854
URL: https://issues.apache.org/jira/browse/FLINK-10854
Project: Flink
Issue Type: Test
Hi to all,
in our ETL we need to call an external (REST) service once a job ends: we
extract informations about accumulators and we update the job status.
However this is only possible if using the CLI client: if we call the job
via the REST API o Web UI (that is very useful to decouple our UI from
ouyangzhe created FLINK-10852:
-
Summary: Decremented number of unfinished producers below 0. This
is most likely a bug in the execution state/intermediate result partition
management
Key: FLINK-10852
URL: https://iss
frank wang created FLINK-10851:
--
Summary: sqlUpdate support complex insert grammar
Key: FLINK-10851
URL: https://issues.apache.org/jira/browse/FLINK-10851
Project: Flink
Issue Type: Bug
ouyangzhe created FLINK-10850:
-
Summary: Job may hang on FAILING state if taskmanager
updateTaskExecutionState failed
Key: FLINK-10850
URL: https://issues.apache.org/jira/browse/FLINK-10850
Project: Flink
15 matches
Mail list logo