From: | David Steele <david(at)pgmasters(dot)net> |
---|---|
To: | Michael Banck <michael(dot)banck(at)credativ(dot)de> |
Cc: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com>, Bernd Helmle <mailings(at)oopsware(dot)de>, PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: [patch] reorder tablespaces in basebackup tar stream for backup_label |
Date: | 2017-03-14 14:34:28 |
Message-ID: | d53ef502-58aa-3905-b160-9c01bce034af@pgmasters.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 3/4/17 2:20 AM, Michael Paquier wrote:
> On Sat, Mar 4, 2017 at 1:13 AM, Bernd Helmle <mailings(at)oopsware(dot)de> wrote:
>> Ah right, i assumed there must be something, otherwise the comment
>> won't be there ;)
>>
>> We could special case that part to distinguish fetch/stream mode, but i
>> fear that leads to more confusion than it wants to fix. The other
>> option of a separate tar file looks awkward from a usability point of
>> view.
>
> It is possible as well to group the WAL files in a different tarball
> than the main directory and put that at the tail of the list for the
> fetch mode, while the main directory gets at the head. That would
> break things for existing users by the way, and just being able to
> look at the LSN start location before receiving the tar bytes of the
> backup does not seem enough to justify such a breakage.
This thread is stalled and it looks like the patch may not be workable,
at least in the current form.
I will mark this a "Returned with Feedback" on 2017-03-17 unless there
are arguments to the contrary.
Thanks,
--
-David
david(at)pgmasters(dot)net
From | Date | Subject | |
---|---|---|---|
Next Message | Alvaro Herrera | 2017-03-14 14:36:43 | Re: Patch: Optimize memory allocation in function 'bringetbitmap' |
Previous Message | David Steele | 2017-03-14 14:18:47 | Re: Patch: Optimize memory allocation in function 'bringetbitmap' |