I like option 1 the most ("move to flink-core"), however, it would scatter
the type extractor / type information classes accross multiple projects.
Why are we not moving the entire type extractor system into "flink-core"?
There were some thoughts in the past to restructure the maven module
layout:
Are you running a fixed number of iterations or do you use a dynamic
termination criterion?
For fixed iterations, you can get the id of the current iteration ... which
allows you to find out when you are running the last iterations.
Would it be feasible for you to just log these statistics to the
We could also add a link to the documentation into the exception that
explains the behavior.
On Fri, Jun 19, 2015 at 5:52 AM, Chiwan Park wrote:
> +1 for ignoring execute() call with warning.
>
> But I'm concerned for how the user catches the error in program without
> any data sinks.
>
> By the
I don't know whether its a good idea to move a JIRA discussion to the dev@
list.
I agree with Ufuk. The logs are the only way of finding out whether there
is something happening or not.
As I said in the JIRA, I think its totally fine to log this at INFO level.
If a production user has identified t
Recently I was running a job and the only way to tell how far it was and to
make sure that it was still running were the JM logs with the current split
assignments.
I like Fabians suggestion. But as long as we don't have good runtime
monitoring in place there is a valid use case for the logs (whic
The ratio could be directly logged. That would be easier that estimating
the ratio from potentially hundreds of log statements of several data
sources.
On Jun 20, 2015 22:49, "Robert Metzger (JIRA)" wrote:
>
> [
> https://issues.apache.org/jira/browse/FLINK-2252?page=com.atlassian.jira.plugin
Thanks, Ted - great news. :)
@Stephan: Streaming is mentioned, but not really prominent. "... could be
just the ticket for fusing streaming analytics with historical analytics."
On Sat, Jun 20, 2015 at 8:23 PM, Ted Dunning wrote:
> On Sat, Jun 20, 2015 at 12:52 AM, Stephan Ewen wrote:
>
> > T
On Sat, Jun 20, 2015 at 12:52 AM, Stephan Ewen wrote:
> There is no comment section, otherwise
> I would have added that...
>
Email to the author. Better coverage next time.
Thanks you for forwarding this, Ted!
I also found this today and was very happy to see Flink being up in the
list.
They labeled Flink as being mainly strong in iterations. While it is true
that Flink is strong in iterations, it is also particularly strong in
streaming, which was not mentioned. Th
