Re: prevent immature WAL streaming - Mailing list pgsql-hackers

From Kyotaro Horiguchi
Subject Re: prevent immature WAL streaming
Date
Msg-id 20210826.102052.1246520870983523562.horikyota.ntt@gmail.com
Whole thread Raw
In response to Re: prevent immature WAL streaming  ("alvherre@alvh.no-ip.org" <alvherre@alvh.no-ip.org>)
List pgsql-hackers
At Wed, 25 Aug 2021 20:20:04 -0400, "alvherre@alvh.no-ip.org" <alvherre@alvh.no-ip.org> wrote in 
> BTW while going about testing this, I noticed that we forbid
> pg_walfile_name() while in recovery.  That restriction was added by
> commit 370f770c15a4 because ThisTimeLineID was not set correctly during
> recovery.  That was supposed to be fixed by commit 1148e22a82ed, so I
> thought that it should be possible to remove the restriction.  However,
> I did that per the attached patch, but was quickly disappointed because
> ThisTimeLineID seems to remain zero in a standby for reasons that I
> didn't investigate.

On a intermediate node of a cascading replication set, timeline id on
walsender and walrecever can differ and ordinary backends cannot
decide which to use.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center



pgsql-hackers by date:

Previous
From: Masahiko Sawada
Date:
Subject: Re: Failure of subscription tests with topminnow
Next
From: Kyotaro Horiguchi
Date:
Subject: Re: prevent immature WAL streaming