>From: Ranier Vilela <ranier.vf@gmail.com> >Sent: Thursday, February 24, 2022 08:46 >To: Justin Pryzby <pryzby@telsasoft.com> >Cc: ldh@laurent-hasson.com; pgsql-performance@postgresql.org >Subject: Re: An I/O error occurred while sending to the backend (PG 13.4) > >Em qui., 24 de fev. de 2022 às 09:59, Justin Pryzby <mailto:pryzby@telsasoft.com> escreveu: >On Thu, Feb 24, 2022 at 08:50:45AM -0300, Ranier Vilela wrote: >> I can't understand why you are still using 13.4? >> [1] There is a long discussion about the issue with 13.4, the project was >> made to fix a DLL bottleneck. >> >> Why you not use 13.6? > >That other problem (and its fix) were in the windows build environment, and not >an issue in some postgres version. >Yeah, correct. >But I think that it was very clear in the other thread that version 13.4, >on Windows, may have a slowdown, because of the DLL problem. >So it would be better to use the latest available version >that has this specific fix and many others. > >regards, >Ranier Vilela
OK, absolutely. I was thinking about even moving to 14. I know migrations within a release are painless, but my experience with upgrading across releases has also been quite good (short of bugs that were found of course). Any opinion on 14.2?
Of course, 14.2 would be better than 13.6, but I think that there are
chances that this specific problem is not beneficial.
And both 13.6 and 14.2 still suffer from a Windows version specific issue [1].
A solution has been proposed which has not yet been accepted.
But in general terms, you will benefit from adopting 14.2 for sure.
