linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andreas Klauer <Andreas.Klauer@metamorpher.de>
To: Alexander Shenkin <al@shenkin.org>
Cc: linux-raid@vger.kernel.org
Subject: Re: recovering failed raid5
Date: Wed, 16 Nov 2016 12:14:07 +0100	[thread overview]
Message-ID: <20161116111407.GA3348@metamorpher.de> (raw)
In-Reply-To: <194ae464-2d19-cefd-055f-6f4a33676ab2@shenkin.org>

On Wed, Nov 16, 2016 at 09:04:29AM +0000, Alexander Shenkin wrote:
> I'm getting a 'No space left on device error'.  Any thoughts?

It's smaller by 4096 bytes, that's probably not a problem. 
ddrescue seems to have failed to copy 128K of data, 
but that's probably not a big problem either.

Your problem is something else:

> Disk /dev/sdb: 2.7 TiB, 3000592982016 bytes, 5860533168 sectors
> Sector size (logical/physical): 512 bytes / 512 bytes
> Disklabel type: gpt

> Disk /dev/sdc: 2.7 TiB, 3000592977920 bytes, 732566645 sectors
> Sector size (logical/physical): 4096 bytes / 4096 bytes
> Disklabel type: dos

The physical sector size is different.

Unfortunately GPT partition scheme still depends on sector size and is 
inherently incompatible when dd(rescue)'d to a drive with different size.

In theory it would be possible to ignore this, i.e. interpret GPT 
correctly on a 4K sector drive even if it was created for a 512b drive, 
or vice versa, but Linux is quite strict about standards in this case. 
(If Linux was smarter it would work in Linux but fail for Windows...)

Anyway, you'll have to fixer-upper your GPT partition tables to 4K. 
gdisk has an expert -> recovery section that might be able to do so 
automagically, or you could just manually recreate with the correct 
_byte_ offsets (sector offset will be different).

Your partitions are all MiB aligned so there are no alignment issues.

Once the partition table is fixed you should be able to proceed normally.

Regards
Andreas Klauer

  reply	other threads:[~2016-11-16 11:14 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-27 15:06 recovering failed raid5 Alexander Shenkin
2016-10-27 16:04 ` Andreas Klauer
2016-10-28 12:22   ` Alexander Shenkin
2016-10-28 13:33     ` Andreas Klauer
2016-10-28 21:16       ` Phil Turmel
2016-10-28 23:45         ` Andreas Klauer
2016-10-29  2:52           ` Edward Kuns
2016-10-29  2:53           ` Phil Turmel
2016-10-29  8:46           ` Mikael Abrahamsson
2016-10-29 10:29       ` Roman Mamedov
2016-10-29 12:02         ` Andreas Klauer
2016-10-30 16:18           ` Phil Turmel
2016-10-28 13:36     ` Robin Hill
2016-10-31 10:44       ` Alexander Shenkin
2016-10-31 11:09         ` Andreas Klauer
2016-10-31 15:19         ` Robin Hill
2016-10-31 16:26         ` Wols Lists
2016-10-31 16:28       ` Wols Lists
2016-11-16  9:04       ` Alexander Shenkin
2016-11-16 11:14         ` Andreas Klauer [this message]
2016-11-16 13:27           ` Alexander Shenkin
2016-11-16 13:59             ` Andreas Klauer
2016-11-16 15:35         ` Wols Lists
2016-11-16 15:50           ` Alexander Shenkin
2016-11-16 16:38             ` Wols Lists
2017-01-05 12:08               ` Alexander Shenkin
2016-10-31 16:31     ` Wols Lists
2016-10-27 16:26 ` Roman Mamedov
2016-10-27 20:34 ` Robin Hill

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=20161116111407.GA3348@metamorpher.de \
    --to=andreas.klauer@metamorpher.de \
    --cc=al@shenkin.org \
    --cc=linux-raid@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).