Re: Backup throttling

From: Antonin Houska <antonin(dot)houska(at)gmail(dot)com>
To: Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>
Cc: "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Backup throttling
Date: 2014-02-28 08:16:23
Message-ID: 531045D7.9080703@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 02/27/2014 11:04 PM, Alvaro Herrera wrote:
> I pushed this patch with a few further tweaks. In your changes to
> address the above point, you made the suffix mandatory in the
> pg_basebackup -r option. This seemed a strange restriction, so I
> removed it. It seems more user-friendly to me to accept the value as
> being expressed in kilobytes per second without requiring the suffix to
> be there; the 'k' suffix is then also accepted and has no effect. I
> amended the docs to say that also.
>
> If you or others feel strongly about this, we can still tweak it, of
> course.

I'm used to assume the base unit if there's no suffix, but have no
objections against considering kB as the default. I see you adjusted
documentation too.

> I also moved the min/max #defines to replication/basebackup.h, and
> included that file in pg_basebackup.c. This avoids the duplicated
> values. That file is okay to be included there.

I kept in mind that pg_basebackup.c is not linked to the backend, but
you're right, mere inclusion is something else.

> Thanks for your patch, and the numerous reviewers who took part.

Thanks for committing - this is my first patch :-)

// Tony

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Kyotaro HORIGUCHI 2014-02-28 08:55:21 Hot standby doesn't come up on some situation.
Previous Message KONDO Mitsumasa 2014-02-28 08:15:17 Re: What behavior is in this loop?