Re: walsummarizer can get stuck when switching timelines - Mailing list pgsql-hackers

From Robert Haas
Subject Re: walsummarizer can get stuck when switching timelines
Date
Msg-id CA+TgmoZ7-CaTEfP8xN1B7ncB4r1y1FMhPtm5LdwVM3sek6wEug@mail.gmail.com
Whole thread
In response to Re: walsummarizer can get stuck when switching timelines  (Thom Brown <thom@linux.com>)
List pgsql-hackers
On Fri, Jul 17, 2026 at 12:22 PM Thom Brown <thom@linux.com> wrote:
> It's erroring out, but shouldn't it check for 000000020000000000000003
> before doing so? It wasn't removed, just renamed.

Actually, neither is the case. 000000010000000000000003 was not
archived. 000000020000000000000003 is a different file with partially
identical contents. But the fact that the existing error message here
is not quite right is not a problem this patch should be trying to
solve.

--
Robert Haas
EDB: http://www.enterprisedb.com



pgsql-hackers by date:

Previous
From: Scott Ray
Date:
Subject: Re: Report oldest xmin source when autovacuum cannot remove tuples
Next
From: Sami Imseih
Date:
Subject: Re: Report index currently being vacuumed in pg_stat_progress_vacuum