Hi Chesnay,
Thanks for the proposal. +1 for make the distribution thinner.
Meanwhile, it would be useful to have all the peripheral libraries/jars
hosted somewhere so users can download them from a centralized place. We
can also encourage the community to contribute their libraries, such as
conne
+1. Great idea!
On Mon, Jan 28, 2019 at 12:02 AM Shaoxuan Wang wrote:
> +1. Looking forward to it.
>
> On Thu, Jan 24, 2019 at 7:41 PM Robert Metzger
> wrote:
>
> > Thanks for your response.
> >
> > You are right, I'm proposing "user...@flink.apache.org" as the mailing
> > list's name!
> >
> >
Thanks all for the replies.
Though the current implementation in blink can handle some cases of the problem,
the solution is not straight forward and efficient enough and can
result in unnecessary
restarting of terminated upstream task. The following word count job
with parallelism 2
can demonstr
+1. Looking forward to it.
On Thu, Jan 24, 2019 at 7:41 PM Robert Metzger wrote:
> Thanks for your response.
>
> You are right, I'm proposing "user...@flink.apache.org" as the mailing
> list's name!
>
> On Thu, Jan 24, 2019 at 12:37 PM Tzu-Li (Gordon) Tai
> wrote:
>
> > Hi Robert,
> >
> > Thank
IMHO it would be a pity if we lost the Git history of changes made to the
Blink repository since the time that it was forked from Flink.
If the Blink repository is a true fork, i.e., branch at a certain
time/commit id, from the Flink repository, then there is no problem. The
entire Blink repositor
