+1 for this. We can wait to gather a broader consensus for certain rare cases where we add new APIs or public changes but if the reviewer's original review comments were addressed then it should be ok to go ahead and merge post a timeout value.
On Tue, Jan 31, 2023 at 2:51 PM Uma Maheswara Rao Gangumalla < umamah...@apache.org> wrote: > Dear Ozone Devs, > > This has been brought up by Ritesh in community sync. I thought we could > discuss this more broadly and set up some path forwarding options. > > With my past experience, when a reviewer does not respond for a long time > and if another reviewer helped to review, we will just give a few days' > wait for the old reviewer to confirm his comments. If no further responses > from the old reviewer, we used to move ahead. > > I think we can follow the similar approach: > > When the reviewer does not respond for a long time (say 15days? ): > The Author can freely ping other reviewers to help if no one else > reviewed it yet. > If the other reviewer is a committer and got +1 from him, then we can > give a 3-7days time window for the old reviewer to check as a final > friendly reminder. If still no responses, then we can just move ahead for > commit based new committer reviewer's +1. > > Thoughts? > > Regards, > Uma >
