Re: Updated backup APIs for non-exclusive backups

From: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
To: Magnus Hagander <magnus(at)hagander(dot)net>
Cc: Marco Nenciarini <marco(dot)nenciarini(at)2ndquadrant(dot)it>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Updated backup APIs for non-exclusive backups
Date: 2016-04-01 04:47:45
Message-ID: CAA4eK1JjbY1pZmCyX_1OeCH4jpE=7UPoZw9QQ3bOYOY=nvhMhw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sat, Mar 19, 2016 at 5:45 PM, Magnus Hagander <magnus(at)hagander(dot)net>
wrote:

> On Wed, Mar 2, 2016 at 6:49 PM, Marco Nenciarini <
> marco(dot)nenciarini(at)2ndquadrant(dot)it> wrote:
>
>>
>> I've attached an updated patch, which is rebased on current master and
> includes the oid fix.
>
>

+ <entry>Finish performing exclusive on-line backup (restricted to
superusers or replication roles)</entry>

+ </row>

+ <row>

+ <entry>

+ <literal><function>pg_stop_backup(<parameter>exclusive</>
<type>boolean</>)</function></literal>

+ </entry>

+ <entry><type>setof record</type></entry>

+ <entry>Finish performing exclusive or non-exclusive on-line backup
(restricted to superusers or replication roles)</entry>

Isn't it better to indicate that user needs to form a backup_label and
tablespace_map file from the output of this API and those needs to be
dropped into data directory?

Also, I think below part of documentation for pg_start_backup() needs to be
modified:

<para>

<function>pg_start_backup</> accepts an

arbitrary user-defined label for the backup. (Typically this would be

the name under which the backup dump file will be stored.) The function

writes a backup label file (<filename>backup_label</>) and, if there

are any links in the <filename>pg_tblspc/</> directory, a tablespace map

file (<filename>tablespace_map</>) into the database cluster's data

directory, performs a checkpoint, and then returns the backup's starting

transaction log location as text. The user can ignore this result
value,

but it is provided in case it is useful.

Similarly, there is a description for pg_stop_backup which needs to be
modified.

CREATE OR REPLACE FUNCTION

- pg_start_backup(label text, fast boolean DEFAULT false)

+ pg_start_backup(label text, fast boolean DEFAULT false, exclusive
boolean DEFAULT true)

RETURNS pg_lsn STRICT VOLATILE LANGUAGE internal AS 'pg_start_backup';

One thing, that might be slightly inconvenient for user is if he or she
wants to use this API for non-exclusive backups then, they need to pass the
value of second parameter as well which doesn't seem to be a big issue.

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Craig Ringer 2016-04-01 04:47:56 Re: Re: [COMMITTERS] pgsql: Enable logical slots to follow timeline switches
Previous Message Tom Lane 2016-04-01 04:30:31 Re: Incorrect format in error message