From: Andrei Borzenkov <arvidjaar@gmail.com>
To: Chris Murphy <lists@colorremedies.com>
Cc: "Austin S. Hemmelgarn" <ahferroin7@gmail.com>,
Hugo Mills <hugo@carfax.org.uk>,
Zygo Blaxell <ce3g8jdj@umail.furryterror.org>,
kreijack@inwind.it, Roman Mamedov <rm@romanrm.net>,
Btrfs BTRFS <linux-btrfs@vger.kernel.org>
Subject: Re: Adventures in btrfs raid5 disk recovery
Date: Sun, 26 Jun 2016 10:54:16 +0300 [thread overview]
Message-ID: <576F8A28.7060808@gmail.com> (raw)
In-Reply-To: <CAJCQCtQhRbL_AWpspudPZzkkGkHT3O9T0QL94Of5LD8agmonwg@mail.gmail.com>
26.06.2016 00:52, Chris Murphy пишет:
> Interestingly enough, so far I'm finding with full stripe writes, i.e.
> 3x raid5, exactly 128KiB data writes, devid 3 is always parity. This
> is raid4.
That's not what code suggests and what I see in practice - parity seems
to be distributed across all disks; each new 128KiB file (extent) has
parity on new disk. At least as long as we can trust btrfs-map-logical
to always show parity as "mirror 2".
Do you see consecutive full stripes in your tests? Or how do you
determine which devid has parity for a given full stripe? This
information is not actually stored anywhere, it is computed based on
block group geometry and logical stripe offset.
P.S. usage of "stripe" to mean "stripe element" actually adds to
confusion when reading code :)
next prev parent reply other threads:[~2016-06-26 7:54 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-20 3:44 Adventures in btrfs raid5 disk recovery Zygo Blaxell
2016-06-20 18:13 ` Roman Mamedov
2016-06-20 19:11 ` Zygo Blaxell
2016-06-20 19:30 ` Chris Murphy
2016-06-20 20:40 ` Zygo Blaxell
2016-06-20 21:27 ` Chris Murphy
2016-06-21 1:55 ` Zygo Blaxell
2016-06-21 3:53 ` Zygo Blaxell
2016-06-22 17:14 ` Chris Murphy
2016-06-22 20:35 ` Zygo Blaxell
2016-06-23 19:32 ` Goffredo Baroncelli
2016-06-24 0:26 ` Chris Murphy
2016-06-24 1:47 ` Zygo Blaxell
2016-06-24 4:02 ` Andrei Borzenkov
2016-06-24 8:50 ` Hugo Mills
2016-06-24 9:52 ` Andrei Borzenkov
2016-06-24 10:16 ` Hugo Mills
2016-06-24 10:19 ` Andrei Borzenkov
2016-06-24 10:59 ` Hugo Mills
2016-06-24 11:36 ` Austin S. Hemmelgarn
2016-06-24 17:40 ` Chris Murphy
2016-06-24 18:06 ` Zygo Blaxell
2016-06-24 17:06 ` Chris Murphy
2016-06-24 17:21 ` Andrei Borzenkov
2016-06-24 17:52 ` Chris Murphy
2016-06-24 18:19 ` Austin S. Hemmelgarn
2016-06-25 16:44 ` Chris Murphy
2016-06-25 21:52 ` Chris Murphy
2016-06-26 7:54 ` Andrei Borzenkov [this message]
2016-06-26 15:03 ` Duncan
2016-06-26 19:30 ` Chris Murphy
2016-06-26 19:52 ` Zygo Blaxell
2016-06-27 11:21 ` Austin S. Hemmelgarn
2016-06-27 16:17 ` Chris Murphy
2016-06-27 20:54 ` Chris Murphy
2016-06-27 21:02 ` Henk Slager
2016-06-27 21:57 ` Zygo Blaxell
2016-06-27 22:30 ` Chris Murphy
2016-06-28 1:52 ` Zygo Blaxell
2016-06-28 2:39 ` Chris Murphy
2016-06-28 3:17 ` Zygo Blaxell
2016-06-28 11:23 ` Austin S. Hemmelgarn
2016-06-28 12:05 ` Austin S. Hemmelgarn
2016-06-28 12:14 ` Steven Haigh
2016-06-28 12:25 ` Austin S. Hemmelgarn
2016-06-28 16:40 ` Steven Haigh
2016-06-28 18:01 ` Chris Murphy
2016-06-28 18:17 ` Steven Haigh
2016-07-05 23:05 ` Chris Murphy
2016-07-06 11:51 ` Austin S. Hemmelgarn
2016-07-06 16:43 ` Chris Murphy
2016-07-06 17:18 ` Austin S. Hemmelgarn
2016-07-06 18:45 ` Chris Murphy
2016-07-06 19:15 ` Austin S. Hemmelgarn
2016-07-06 21:01 ` Chris Murphy
2016-06-24 16:52 ` Chris Murphy
2016-06-24 16:56 ` Hugo Mills
2016-06-24 16:39 ` Zygo Blaxell
2016-06-24 1:36 ` Zygo Blaxell
2016-06-23 23:37 ` Chris Murphy
2016-06-24 2:07 ` Zygo Blaxell
2016-06-24 5:20 ` Chris Murphy
2016-06-24 10:16 ` Andrei Borzenkov
2016-06-24 17:33 ` Chris Murphy
2016-06-24 11:24 ` Austin S. Hemmelgarn
2016-06-24 16:32 ` Zygo Blaxell
2016-06-24 2:17 ` Zygo Blaxell
2016-06-22 4:06 ` Adventures in btrfs raid5 disk recovery - update Zygo Blaxell
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=576F8A28.7060808@gmail.com \
--to=arvidjaar@gmail.com \
--cc=ahferroin7@gmail.com \
--cc=ce3g8jdj@umail.furryterror.org \
--cc=hugo@carfax.org.uk \
--cc=kreijack@inwind.it \
--cc=linux-btrfs@vger.kernel.org \
--cc=lists@colorremedies.com \
--cc=rm@romanrm.net \
/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;
as well as URLs for NNTP newsgroup(s).