Soumyadeep Chakraborty
Soumyadeep Chakraborty
Thank you for the detailed feedback! I will address it.
I rebased the branch against latest master and then made a separate commit (to be squashed later) to address the feedback so far. I had a few questions and comments...
@DimCitus w/ @jchampio Addressed much of the feedback you had for us. We haven't yet done anything to change the retry logic on the client side `pg_autoctl set` process (the...
Hey @DimCitus , Thanks for the detailed steps! I have a somewhat alternate idea. what if the user had a way to provide `pgautofailover_replicator` passwords if `--auth` is password based...
> About your next idea and the pre-existing partial configuration, I believe that the way our code is written that's already possible. If you provision a pg_autoctl configuration before running...
> I'm not sure about the level of complexity for this idea as far as end-users are concerned. On the other hand it's an advanced command, and could be documented/introduced...
> Another approach here that leverages flex/bison and thus is probably more robust... [dgkimura/postgres@02d4bb3](https://github.com/dgkimura/postgres/commit/02d4bb300a384db62ef5766db5fe105c7bd6ef1a) I added minor feedback to that. That approach looks good! Lets implement for the the other...
Hi @maksm90 , apologies for the slow response. Thank you very much for raising the PR! It is on our list of resource queue related improvements. We will look at...
@Aegeaner (cc @adam8157 ) Can we restart work on this soon?
This is ready for review. Refactoring the MotionConn can be done as a separate PR.