From: | Alexander Kukushkin <cyberdemn(at)gmail(dot)com> |
---|---|
To: | pgsql-general(at)lists(dot)postgresql(dot)org |
Subject: | Re: Patroni, slots, and expiring WALs |
Date: | 2023-03-28 15:08:38 |
Message-ID: | CAFh8B=nf_Xgc-YaLE3vo=T6c9OqcnMJNa-stMh8R_jofeopidQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Hi,
On Tue, 28 Mar 2023 at 16:55, Peter J. Holzer <hjp-pgsql(at)hjp(dot)at> wrote:
>
> However, when we took down one node for about two hours for some tests
> recently (with some moderate traffic on the remaining node), the replica
> didn't catch up after being restarted and inspection of the logs showed
> that it was trying to get WALs which had already been deleted.
>
It is impossible to know whether it is down temporarily or gone forever,
therefore the slot is removed as soon as the member key expires (after TTL).
> So apparently, `use_slots: true` isn't enough. What else do I have to
> configure? (I know about wal_keep_size, but it was my understanding that
> this isn't needed when slots are used)
>
The best option is to configure continuous archiving and PITR. Backups are
always important.
The second option - you can put all member names into permanent slots
configuration (using patronictl edit-config):
slots:
nodename1:
type: physical
nodename2:
type: physical
nodename3:
type: physical
This way slots representing these members will not be removed.
Regards,
--
Alexander Kukushkin
From | Date | Subject | |
---|---|---|---|
Next Message | Peter J. Holzer | 2023-03-28 15:09:25 | Re: Patroni, slots, and expiring WALs |
Previous Message | Peter J. Holzer | 2023-03-28 15:08:09 | Re: Cluster table based on grand parent? |