In this branch: https://github.com/zentol/flink/tree/new-iterations you
can find a more fine-grained fix for chaining with
iterations. relevant commit: ac2305d9589a5c6ab9e94d04c870fba52716d695
On 13.10.2016 23:11, Chesnay Schepler wrote:
The chaining code is definitely related, I also have a pretty clear
idea how to fix it.
The odd thing is that the Java API doesn't catch this type mismatch;
the date types are
known when the plan is generated. This kind of error shouldn't even
On 13.10.2016 21:15, Geoffrey Mon wrote:
Thank you very much. Disabling chaining with the Python API allows my
actual script to run properly. The division by zero must be an issue
the job that I posted on gist.
Does that mean that the issue must be in the chaining part of the API?
Chaining from the way I understand it is an important optimization that
would be important for the performance comparison I wish to make in my
On Thu, Oct 13, 2016 at 9:11 AM Chesnay Schepler <ches...@apache.org>
A temporary work around appears to be disabling chaining, which you can
do by commenting out L215 "self._find_chains()" in Environment.py.
Note that you then run into a division by zero error, but i can't tell
whether that is a problem of the job or not.
On 13.10.2016 13:41, Chesnay Schepler wrote:
I was able to reproduce the error and will look into it in more detail
On 12.10.2016 23:09, Geoffrey Mon wrote:
Has anyone had a chance to look into this? I am currently working
problem but I have minimal understanding of how the internal Flink
API works; any expertise would be greatly appreciated.
Thank you very much!
On Sat, Oct 8, 2016 at 1:27 PM Geoffrey Mon <geof...@gmail.com>
Heh, I have discovered that if I do not restart Flink after
original problematic script, then similar issues will manifest
in other otherwise working scripts. I haven't been able to
narrow down the problem, but I promise this new script will have a
ClassCastException that is completely reproducible. :)
On Wed, Sep 28, 2016 at 9:16 AM Chesnay Schepler
this one works for me as well :D
On 28.09.2016 05:38, Geoffrey Mon wrote:
Thank you for your help. After receiving your message I
version of Flink completely, and both the NullPointerException
the TODO and the ClassCastException with the join operation went
Previously, I had been only recompiling the modules of Flink
changed to save time using "mvn clean install -pl :module" and
that may have been causing some of my issues.
Now, the problem is more clear: when a specific group reduce
research project plan file is used within an iteration, I get a
Caused by: java.lang.ClassCastException:
org.apache.flink.api.java.tuple.Tuple2 cannot be cast to [B
I'm not sure why this is causing an exception, and I would greatly
appreciate any assistance. I've revised the barebones error-causing
file to focus on this new error source:
The group reduce function in question seems to work just fine
iterations. I have organized the commits and pushed to a new
it easier to test and hopefully review soon:
On Mon, Sep 26, 2016 at 6:32 AM Chesnay Schepler
i could not reproduce this issue with the commits and plan you
I tried out both the FLINK-4098 and bulk-iterations branches (and
reverted back to the specified commits) and built Flink from
Could you double check that the code you provided produces the
Also, which OS/python version are you using?
On 20.09.2016 11:13, Chesnay Schepler wrote:
I'll try to take a look this week.
On 20.09.2016 02:38, Geoffrey Mon wrote:
I have recently been working on adding bulk iterations to the
Flink in order to facilitate a research project I am working on.
current changes can be seen in this GitHub diff:
This implementation seems to work for, at least, simple
incrementing numbers in a data set. However, with the
required for my project, I get an exception
[B cannot be cast to org.apache.flink.api.java.tuple.Tuple"
deserializers called by
I've created the following simplified Python plan by stripping
research project code to the problem-causing parts:
I have been working on this issue but I don't have any ideas on
be the problem. Perhaps someone more knowledgeable about the
the Python API could kindly help?
Thank you very much.