Re: pg_recvlogical broken in back branches

From: Noah Misch <noah(at)leadboat(dot)com>
To: Michael Paquier <michael(at)paquier(dot)xyz>, magnus(at)hagander(dot)net
Cc: Euler Taveira <euler(at)timbira(dot)com(dot)br>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: pg_recvlogical broken in back branches
Date: 2018-04-22 21:55:51
Message-ID: 20180422215551.GB2676194@rfd.leadboat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Apr 17, 2018 at 03:38:13PM +0900, Michael Paquier wrote:
> On Tue, Apr 17, 2018 at 03:01:33AM -0300, Euler Taveira wrote:
> > A proposed fix is attached. It should be applied to 9.4, 9.5, 9.6, and
> > 10. (Although, client version 10 can connect to server version 10,
> > client version 10 can't connect to server version 9.6.)
> >
> > Comments?
>
> The exact same fix has already applied on all stable branches:
> - af5fbb1286 -> REL9_4_STABLE
> - 24ff0fe877 -> REL9_5_STABLE
> - 59743deca9 -> REL9_6_STABLE
> - e7d3a37d99 -> REL_10_STABLE
> - 8d2814f274 -> master

That change is testing the wrong variable. I plan to repair it as attached.
I ran check-world with the following and found no similar defects:

--- a/src/interfaces/libpq/fe-connect.c
+++ b/src/interfaces/libpq/fe-connect.c
@@ -6106,4 +6106,5 @@ int
PQserverVersion(const PGconn *conn)
{
+ Assert(conn);
if (!conn)
return 0;

Attachment Content-Type Size
recvlogical-fix-version-test-v1.patch text/plain 1.1 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2018-04-22 22:49:05 Re: pg_recvlogical broken in back branches
Previous Message Andrew Gierth 2018-04-22 20:25:06 Re: Toast issues with OldestXmin going backwards