From: Dave Chinner <david@fromorbit.com>
To: L A Walsh <xfs@tlinx.org>
Cc: linux-xfs <linux-xfs@vger.kernel.org>
Subject: Re: RFD: question/probe regarding adding dump-levels A-Z after 0-9 for xfs{dump,restore}
Date: Tue, 11 Dec 2018 08:55:07 +1100 [thread overview]
Message-ID: <20181210215507.GD6311@dastard> (raw)
In-Reply-To: <5C0A70FD.6080806@tlinx.org>
On Fri, Dec 07, 2018 at 05:09:17AM -0800, L A Walsh wrote:
> How difficult would it be to allow 26 more dump levels
> in xfs dump/restore? Theoretically anyone using 0-9
> would see no change. Would there be some "problem" or
> "gotcha" in adding in what would _seem_ to be a simple
> change?
Everything _seems_ simple with xfsdump. e.g. the inventory
session headers supports up to 255 levels in it's on-disk format,
but the xfsdump command is limited to just 10 levels by a separate
"max level" define.
IOWs, on the surface it _seems_ simple to change it, but I don't
know all the places that the dump level is stored, nor what would
happen if an older xfsdump/xfsrestore binary tripped over an
inventory with a dump level larger than they were compiled to
handle.
I also don't know if there are scalability problems with resolving
the contents of incremental dump levels. Certainly restoring from 30
incremental dumps is more onerous than 6-7, but that's just time.
What I don't know is how the invenetory scales to more than 10
incremental and whether there's some exponential algorithm in there
that falls apart. So there's a lot of testing that would be needed
to validate what looks like a 1 line change to the code.
So, yes, technically we could increase the dump level, but there's a
*lot* of verification work after doing so and it's not a risk-free
modification.
FWIW, the idea behind 10 dump levels is that it is enough for a
full dump every week w/ an incremental every night of the week until
the next full dump is done on the weekend. What's the backup plan
you want to use/need that requires a larger number of incrementals?
Knowing what you want more dump levels for helps us understand how
it would be used and what really needs to be testedi first....
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
prev parent reply other threads:[~2018-12-10 21:55 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-07 13:09 RFD: question/probe regarding adding dump-levels A-Z after 0-9 for xfs{dump,restore} L A Walsh
2018-12-10 21:55 ` Dave Chinner [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181210215507.GD6311@dastard \
--to=david@fromorbit.com \
--cc=linux-xfs@vger.kernel.org \
--cc=xfs@tlinx.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox