On Fri, Jun 28, 2024 at 6:30 PM Bertrand Drouvot <bertranddrouvot...@gmail.com> wrote: > > On Fri, Jun 28, 2024 at 03:15:22PM +0530, Amit Kapila wrote: > > On Fri, Jun 28, 2024 at 12:55 PM Peter Smith <smithpb2...@gmail.com> wrote: > > > > > > > I don't know whether your assumption is correct. AFAICS, those two > > lines should be together. Let us ee if Bertrand remembers anything? > > > > IIRC the WalSndWaitForWal() call has been moved to ensure that we can > determine > the timeline accurately. >
This part is understandable but I don't understand the part of the comment (This is needed to determine am_cascading_walsender accurately ..) atop a call to WalSndWaitForWal(). The am_cascading_walsender is determined based on the results of RecoveryInProgress(). Can the wait for WAL by using WalSndWaitForWal() change the result of RecoveryInProgress()? -- With Regards, Amit Kapila.