GaoLun created FLINK-2884:
-
Summary: Apply JMH on HashVsSortMiniBenchmark class.
Key: FLINK-2884
URL: https://issues.apache.org/jira/browse/FLINK-2884
Project: Flink
Issue Type: Sub-task
Co
Till Rohrmann created FLINK-2883:
Summary: Combinable reduce produces wrong result
Key: FLINK-2883
URL: https://issues.apache.org/jira/browse/FLINK-2883
Project: Flink
Issue Type: Bug
Aff
Is there guidance for configuring Flink on large clusters? I have recently
been working to benchmark some algorithms on and test AWS. I had no issues
running on a 16 node cluster but when moving to 64 nodes the JobManager
struggled mightily. It did not look to be parallelizing its workload. I was
i
Greg Hogan created FLINK-2882:
-
Summary: Improve performance of string conversions
Key: FLINK-2882
URL: https://issues.apache.org/jira/browse/FLINK-2882
Project: Flink
Issue Type: Improvement
Greg Hogan created FLINK-2881:
-
Summary: WebRuntimeMonitor overly restrictive file patterns
Key: FLINK-2881
URL: https://issues.apache.org/jira/browse/FLINK-2881
Project: Flink
Issue Type: Bug
Hi!
I am not sure what PARTITION_RANDOM is about, but FORWARD and
PARTITION_FORCED_REBALANCE
are the same to the output emitter, but look different to the
Optimizer (PARTITION_FORCED_REBALANCE
cannot be pushed down).
Stephan
On Tue, Oct 20, 2015 at 6:39 PM, Greg Hogan wrote:
> Looking at org.
Looking at org.apache.flink.runtime.operators.shipping.OutputEmitter,
shipping strategies FORWARD, PARTITION_RANDOM, and
PARTITION_FORCED_REBALANCE all call a local round-robin partitioning
function. I'd like to patch this so that the round-robin count starts at
the local task index, but shouldn't
Yes, the G1 core dumps quite often for streaming users. Seems to be
too buggy as a default GC.
On Tue, Oct 20, 2015 at 6:30 PM, Stephan Ewen wrote:
> Looks like a JVM bug.
>
> BTW: We may want to downgrade from G1 as the default GC in streaming mode
> to CMS. In the current Java 8 versions, G1 is
Looks like a JVM bug.
BTW: We may want to downgrade from G1 as the default GC in streaming mode
to CMS. In the current Java 8 versions, G1 is still causing core dumps once
in a while...
Stephan
On Tue, Oct 20, 2015 at 6:23 PM, Maximilian Michels wrote:
> Looks like either a Surefire bug or co
Looks like either a Surefire bug or corrupt memory. Haven't seen this before.
On Tue, Oct 20, 2015 at 6:18 PM, Matthias J. Sax wrote:
> I never saw something like this before... Travis hick up? Can be
> ignored? Or severe issues?
>
> https://travis-ci.org/mjsax/flink/jobs/86431071
>
> -Matthias
>
Stephan Ewen created FLINK-2880:
---
Summary: Allow DeserializationSchema to throw IOException in
deserialization method
Key: FLINK-2880
URL: https://issues.apache.org/jira/browse/FLINK-2880
Project: Flink
I never saw something like this before... Travis hick up? Can be
ignored? Or severe issues?
https://travis-ci.org/mjsax/flink/jobs/86431071
-Matthias
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> SUREFIRE-859: # SIGSEGV (0xb) at pc=0x7f8e02dd96b2, pid=21578,
I'm a little less excited about this. You might not be aware but, for
a large portion of the source code, we already follow the Google style
guide. The main changes will be tabs->spaces and 80/100 characters
line limit.
Out of curiosity, I ran the official Google Style Checkstyle
configuration to
Nikolaas Steenbergen created FLINK-2879:
---
Summary: Links in documentation are broken
Key: FLINK-2879
URL: https://issues.apache.org/jira/browse/FLINK-2879
Project: Flink
Issue Type: Bug
Maximilian Michels created FLINK-2878:
-
Summary: JobManager warns: Unexpected leader address pattern
Key: FLINK-2878
URL: https://issues.apache.org/jira/browse/FLINK-2878
Project: Flink
I
Aljoscha Krettek created FLINK-2877:
---
Summary: Move Streaming API out of Staging
Key: FLINK-2877
URL: https://issues.apache.org/jira/browse/FLINK-2877
Project: Flink
Issue Type: Improvement
1) yes. Been dancing this issue for a while. Let's pull the trigger. Did
the exercise with Tachyon while back and did help readability and
homogeneity of code.
2) +1 for Google Java style with documented exceptions and explanation on
why.
On Tuesday, October 20, 2015, Ufuk Celebi wrote:
> DISCL
Greg Hogan created FLINK-2876:
-
Summary: Minutiae
Key: FLINK-2876
URL: https://issues.apache.org/jira/browse/FLINK-2876
Project: Flink
Issue Type: Improvement
Components: Documentation
I am in favor of Google vanilla code style.
As far as I followed the discussion there will be no style that
everybody loves, but most people agree that there should be a unique
style. Thus, adjusting Google style does not give any benefits.
-Matthias
On 10/20/2015 03:36 PM, Stephan Ewen wrote:
+1 for introducing a stricter style guide and starting with the Google
style.
Should we have a separate discussion whether we take the Google style guide
vanilla, or whether we make slight adjustments?
On Tue, Oct 20, 2015 at 3:26 PM, Till Rohrmann wrote:
> That's how I've understood Ufuk's mai
That's how I've understood Ufuk's mail. Everyone should also be aware that
the Google code style limits the number characters per line to either 80 or
100. But I guess that everyone will read it himself.
On Tue, Oct 20, 2015 at 3:15 PM, Stephan Ewen wrote:
> Just checking: Do we take Google's st
Just checking: Do we take Google's style guide as is, including spaces
instead of tabs? I like the spaces, but that will make things hard...
On Tue, Oct 20, 2015 at 3:01 PM, Gyula Fóra wrote:
> +1 for both :)
>
> Till Rohrmann ezt írta (időpont: 2015. okt. 20., K,
> 14:58):
>
> > I like the ide
Ufuk Celebi created FLINK-2875:
--
Summary: Add createRemoveEnvironment variant for HA
Key: FLINK-2875
URL: https://issues.apache.org/jira/browse/FLINK-2875
Project: Flink
Issue Type: Improvement
+1 for both :)
Till Rohrmann ezt írta (időpont: 2015. okt. 20., K,
14:58):
> I like the idea to have a bit stricter code style which will increase code
> maintainability and makes it easier for people to go through the code.
> Furthermore, it will relieve us from code style comments while review
I like the idea to have a bit stricter code style which will increase code
maintainability and makes it easier for people to go through the code.
Furthermore, it will relieve us from code style comments while reviewing
PRs which can be quite cumbersome.
Personally, I like the Google code style. Th
+1 for both
As we are planning to restructure the maven projects at the point that
breaks the PRs anyway, so going on step further at this point in time is
reasonable for me.
On Tue, Oct 20, 2015 at 2:37 PM, Matthias J. Sax wrote:
> big +1 for both!
>
> On 10/20/2015 02:31 PM, Ufuk Celebi wrote
Ulf Karlsson created FLINK-2874:
---
Summary: Certain Avro generated getters/setters not recognized
Key: FLINK-2874
URL: https://issues.apache.org/jira/browse/FLINK-2874
Project: Flink
Issue Type:
big +1 for both!
On 10/20/2015 02:31 PM, Ufuk Celebi wrote:
> DISCLAIMER: This is not my personal idea, but a community discussion from
> some time ago. Don't kill the messenger.
>
> In March we were discussing issues with heterogeneity of the code [1]. The
> summary is that we had a consensus to
DISCLAIMER: This is not my personal idea, but a community discussion from
some time ago. Don't kill the messenger.
In March we were discussing issues with heterogeneity of the code [1]. The
summary is that we had a consensus to enforce a stricter code style on our
Java code base in order to make i
Yes, that sounds good to me.
Implement support for generic range partitioning first and go for the
non-range-equally-splittable cases later.
Best, Fabian
2015-10-20 5:21 GMT+02:00 Li, Chengxiang :
> Thanks a lot for the comments, Fabian. I agree with you on the plan
> mostly, just add some more
Till Rohrmann created FLINK-2873:
Summary: WebRuntimeMonitor does not start when multiple log files
are present
Key: FLINK-2873
URL: https://issues.apache.org/jira/browse/FLINK-2873
Project: Flink
An example can be found here:
Code to read JSON from line:
https://github.com/dbs-leipzig/gradoop/blob/master/gradoop-flink/src/main/java/org/gradoop/io/json/JsonReader.java#L57
How it is called:
https://github.com/dbs-leipzig/gradoop/blob/master/gradoop-flink/src/main/java/org/gradoop/model/im
The easiest way is to have each json object in a single line so you could
use the env.readTextFile method..
On 20 Oct 2015 09:03, "Papp, Stefan" wrote:
> Hi,
>
>
> I want to process data with JSON. Meaning, I have to receive JSON data and
> prepare this data for analytics. In the beginning, we mi
Hi,
I want to process data with JSON. Meaning, I have to receive JSON data and
prepare this data for analytics. In the beginning, we might receive this data
via files, but I assume soon we will switch to a streaming variant.
What is currently the best recommended practice with Flink?
Thank yo
34 matches
Mail list logo
