MIME-Version: 1.0
References: 
 <CAGXsc+ZmBoLs3Mz=G-Bdm4JJG+fH1NpHfR3qVJVwW4eBKWwStQ@mail.gmail.com>
In-Reply-To: 
 <CAGXsc+ZmBoLs3Mz=G-Bdm4JJG+fH1NpHfR3qVJVwW4eBKWwStQ@mail.gmail.com>
From: Emond Papegaaij <emond.papegaaij@gmail.com>
Date: Tue, 12 May 2026 12:49:31 +0200
Message-ID: 
 <CAGXsc+bBbfW7qQ+2JJ4SY7xZDifrz329cJcxHaoG+kpfGqBJaQ@mail.gmail.com>
Subject: Re: Primary node detection race at clean startup
To: pgpool-hackers@lists.postgresql.org
Content-Type: multipart/mixed; boundary="000000000000eac79106519c9e26"
Archived-At: 
 <https://www.postgresql.org/message-id/CAGXsc%2BbBbfW7qQ%2B2JJ4SY7xZDifrz329cJcxHaoG%2BkpfGqBJaQ%40mail.gmail.com>
Precedence: bulk

--000000000000eac79106519c9e26
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,

Something was wrong with the attached patch. It is rejected by patch,
probably because of the large context. Attached is a new version that
also works with patch.

Best regards,
Emond Papegaaij

Op di 12 mei 2026 om 10:38 schreef Emond Papegaaij <emond.papegaaij@gmail.c=
om>:

>
> Hi,
>
> In our tests, we've found an issue that can cause all Pgpool nodes to
> report an incorrect 'Role: standby':
> Role                   : standby    =E2=86=90 stale, never updated on thi=
s node
> Backend Role           : primary    =E2=86=90 actual SR-check result
>
> This can happen if all nodes in a watchdog cluster start with a clean
> state at the same time. If the first node is still trying to determine
> the primary database, it's primary_node_id is -2. This value is then
> synced to other nodes in the cluster, causing all nodes to report the
> stale state indefinitely. Attached is a patch against 4.7 that should
> fix this.
>
> Note that this analysis was done by Claude Code and it also created
> the patch. The failure on our CI was real though and I think the
> explanation makes sense.
>
> Best regards,
> Emond Papegaaij

--000000000000eac79106519c9e26
Content-Type: application/x-patch;
	name="pgpool-keep-local-primary-when-leader-initial.patch"
Content-Disposition: attachment;
	filename="pgpool-keep-local-primary-when-leader-initial.patch"
Content-Transfer-Encoding: base64
Content-ID: <f_mp2fk08n0>
X-Attachment-Id: f_mp2fk08n0
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--000000000000eac79106519c9e26--