Message-ID: <0ec3ad712a1b8bd63b807091e61e11e5e069b720.camel@cybertec.at>
Subject: Re: Postgre's replica holding huge amount of WAL
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: Mendbayar Alzakhgui <mendbayar.alz@unitel.mn>, Pgsql-admin
	 <pgsql-admin@lists.postgresql.org>
Date: Thu, 15 May 2025 14:27:57 +0200
In-Reply-To: 
 <SEZPR01MB5748AB66F48B9A164728F4A0EA90A@SEZPR01MB5748.apcprd01.prod.exchangelabs.com>
References: 
	<SEZPR01MB5748AB66F48B9A164728F4A0EA90A@SEZPR01MB5748.apcprd01.prod.exchangelabs.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.56.1 (3.56.1-1.fc42) 
MIME-Version: 1.0
Archived-At: 
 <https://www.postgresql.org/message-id/0ec3ad712a1b8bd63b807091e61e11e5e069b720.camel%40cybertec.at>
Precedence: bulk

On Thu, 2025-05-15 at 06:21 +0000, Mendbayar Alzakhgui wrote:
> My PostgreSQL is a cluster that managed by Patroni and etcd. While it has=
 2 replica server
> both of the replica=E2=80=99s are holding huge amount of WAL file (curren=
t 55GB since 29th April).
> I=E2=80=99ve checked the replay delay, wal_keep_size, replication slot, m=
ax_wal_size but all of
> them were fine there=E2=80=99s no slot keeping that wal in replica for su=
re even in master there=E2=80=99s
> no slot holding that WAL, wal_keep_size is 5GB and max_wal_size is 8GB an=
d delay is almost
> 0.02 second almost no delay. But there=E2=80=99s no that size of WAL in m=
aster node.
> What=E2=80=99s the troubleshoot of this case, how to check it?

https://www.cybertec-postgresql.com/en/why-does-my-pg_wal-keep-growing/

Yours,
Laurenz Albe