Hi Piotr, thanks for the proposal, Can you please clarify 1. The scope of the variables added - is it only the last transformation? Do I understand correctly, that chaining does NOT affect this scoping?
2. Is Python API going to be supported as well? Thanks Regards, Roman On Thu, Nov 7, 2024 at 2:52 PM Piotr Nowojski <pnowoj...@apache.org> wrote: > Hi all! > > I would like to open up for discussion a new FLIP-484 [1]. > > Motivation > When running a Flink job that reads data from multiple sources and writes > to multiple sinks, it would be helpful to group metrics from each > source/sink instance together, for example based on the underlying name of > the source/sink table/topic. > > Proposed Changes > This flip proposes to allow users to set custom metric variables to > operators/transformations, that will be later passed on to the metrics and > trace reporters. > > This will allow users to label all of the reported metrics from operators > with for example the table name (SQL), allowing to easily report the number > of records/watermark values per each source/sink table in the same Flink > Job. > > For more information please look into the FLIP-484 [1]. > > I'm looking forward to your thoughts on this. > > Best, > Piotrek > > [1] https://cwiki.apache.org/confluence/x/44yMEw > <https://cwiki.apache.org/confluence/x/4IyMEw> >
