Re: archive status ".ready" files may be created too early - Mailing list pgsql-hackers

On 8/20/21, 4:00 PM, "alvherre@alvh.no-ip.org" <alvherre@alvh.no-ip.org> wrote:
> Attached is v14 which uses a separate spinlock.

This looks good to me.

I was looking at moving the function calls out of the spinlock region.
I don't think the functions are doing anything too expensive, and they
help clean up NotifySegmentsReadyForArchive() quite a bit, but I
understand why it might be against project policy to do something like
that.  It would be easy enough to get rid of the helper functions if
that was concern.

Nathan