Re: [BUG] non archived WAL removed during production crash recovery - Mailing list pgsql-bugs

From Kyotaro Horiguchi
Subject Re: [BUG] non archived WAL removed during production crash recovery
Date
Msg-id 20200410.111454.269125561493324950.horikyota.ntt@gmail.com
Whole thread Raw
In response to Re: [BUG] non archived WAL removed during production crash recovery  (Jehan-Guillaume de Rorthais <jgdr@dalibo.com>)
Responses Re: [BUG] non archived WAL removed during production crash recovery
List pgsql-bugs
At Thu, 9 Apr 2020 18:46:22 +0200, Jehan-Guillaume de Rorthais <jgdr@dalibo.com> wrote in 
> On Wed, 8 Apr 2020 13:58:30 +0200
> Jehan-Guillaume de Rorthais <jgdr@dalibo.com> wrote:
> [...]
> > > May I ask why this new test is added to 011_crash_recovery.pl which is
> > > aimed at testing crash and redo, while we have 002_archiving.pl that
> > > is dedicated to archiving in a more general manner?  
> > 
> > I thought it was a better place because the test happen during crash recovery.
> > 
> > In the meantime, while working on other tests related to $SUBJECT and the
> > current consensus, I was wondering if a new file would be a better place
> > anyway.
> 
> So, 002_archiving.pl deals more with testing recovering on hot standby side
> than archiving. Maybe it could be renamed?

I have the same feeling with Michael.  The test that archives are
created correctly seems to fit the file.  It would be unintentionally
that the file is not exercising archiving so much.

> While discussing this, I created a new file to add some more tests about WAL
> archiving and how recovery deal with them. Please, find the patch in
> attachment. I'll be able to move them elsewhere later, depending on the
> conclusions of this discussion.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center



pgsql-bugs by date:

Previous
From: Kyotaro Horiguchi
Date:
Subject: Re: [BUG] non archived WAL removed during production crash recovery
Next
From: Michael Paquier
Date:
Subject: Re: Loss of replication after simple misconfiguration