Justin Pryzby <pryzby@telsasoft.com> writes:
> On Thu, Apr 01, 2021 at 11:39:33AM -0400, Tom Lane wrote:
>> Will fix, thanks for the report!
> Yes, thank you both. I've run into this recently but for some reason I thought
> it was fixed. It probably also explains this one from 2014.
> https://www.postgresql.org/message-id/20141206061151.GA725@telsasoft.com
Yeah, this does look suspiciously like it explains some past reports
that we failed to reproduce, likely because it didn't occur to us that
reading the file from a non-seekable source would make a difference.
Thanks to Sergey for beating me over the head till I didn't dismiss
it anymore ;-)
regards, tom lane