On Sun, Jun 30, 2024 at 2:40 PM Tom Lane <t...@sss.pgh.pa.us> wrote: > In hopes of moving things along as we approach the v18 branch, > I went ahead and pushed Kuroda-san's patches (with a bit of > further editorialization). AFAICS that allows closing out > the concerns raised by Noah, so I've marked that open item > done. However, I added a new open item about how the > 040_pg_createsubscriber.pl test is slow and still unstable.
This open item is now the only open item. It seems to have been open for a month with no response from Peter which is, from my point of view, far from ideal. However, another thing that is not ideal is that we've been using the same thread to discuss every issue related to this patch for 2.5 years. The thread spans hundreds of messages and it is by no means obvious to what extent the messages posted after this one addressed the underlying concern. Perhaps it would have been an idea to start new threads when we started discussing post-commit issues, instead of just tagging onto the same one. But that said, I see no commits in the commit history which purport to improve performance, so I guess the performance is probably still not what you want, though I am not clear on the details. And as far as stability is concerned, I peered through the dense haze of 027_stream_regress-related buildfarm failures for long enough to discover that the stability issues with 040_pg_createsubscriber aren't fixed either, because we have these recent buildfarm reports: https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=olingo&dt=2024-07-26%2016%3A02%3A40 https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=adder&dt=2024-07-26%2009%3A20%3A15 https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=canebrake&dt=2024-07-25%2002%3A39%3A02 https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=tamandua&dt=2024-07-22%2002%3A31%3A32 So, Peter, as the committer responsible for pg_createsubscriber, what are we going to do about this? -- Robert Haas EDB: http://www.enterprisedb.com
