On 3/4/21 12:56 PM, Andres Freund wrote: > Hi, > > On 2021-03-04 11:10:19 -0500, Andrew Dunstan wrote: >> Here's the patch. > Awesome. Will you commit it?
Done > > >> I didn't see a convenient way of handling the pg_recvlogical case, so >> that's unchanged. > Is the problem actually the kill_kill() itself, or just doing > ->kill_kill() without a subsequent ->finish()? Pretty sure it's the kill_kill that causes the awful crash Michael and I have seen. > > But anyway, that seems like a less critical test... right cheers andrew -- Andrew Dunstan EDB: https://www.enterprisedb.com
