All of lore.kernel.org
 help / color / mirror / Atom feed
From: Phil Turmel <philip@turmel.org>
To: Matt Callaghan <matt_callaghan@sympatico.ca>, linux-raid@vger.kernel.org
Subject: Re: mdadm RAID6 "active" with spares and failed disks; need help
Date: Fri, 27 Mar 2015 21:59:38 -0400	[thread overview]
Message-ID: <55160B0A.1090405@turmel.org> (raw)
In-Reply-To: <BLU437-SMTP28725520E940DA3707A62F81090@phx.gbl>

Hi Matt,

On 03/27/2015 07:48 PM, Matt Callaghan wrote:
> Back at it with fresh brain and fresh hardware. (several months ago I
> got part-way through Valentine's ideas but not all the way -- decided to
> get a clean setup before progressing further)
> 
> I have built a new (fresh/clean) server, and compiled+installed the
> latest mdadm v3.3.2.
> The 8x drives from this RAID6 array have also been moved to the new
> temporary server.
> 
> Now of course, in the new server, the device labels are different.
> I need to map the previous "known labels" in the old server (/dev/sdX)
> to the "new labels" in order to get the drive ordering for re-assembly
> right.
> http://www.linuxquestions.org/questions/linux-server-73/mdadm-raid6-active-with-spares-and-failed-disks%3B-need-help-4175530127/

I read through this.  Given all of the destructive actions you took, I
am doubtful you will ever get your data.  Like mounting "readonly".
That gives you a readonly filesystem, but it writes to the device.
Possibly a great deal if there's a journal to replay.  You also trimmed
much useful data with "grep" that probably would help us save you now.

However, in the hope you might have useful data that can be correlated
with current status, start with lsdrv [1].  Paste the output in your
reply with word wrap turned off.  That'll at least give us a correlation
between device name and serial number.

> e.g. before I had:
> {{{
> /dev/sd[nmlpiokj]1
> }}}

FWIW, it is not safe to use square bracket notation when order matters.

> , and now I have:
> {{{
> /dev/sd[abcdefghi]1
> }}}

The linux 'sd' driver has never guaranteed consistent device names.
It's merely an artifact of boot timing that makes it look that way.
Which is why array members have superblocks that record the roles.  You
absolutely *must* have accurate role numbers to get your data back.

Show complete 'mdadm -E' output for all of your member partitions as
they stand now.

> Unfortunately I don't have any smartctl output saved from the previous
> server and I can't find a way to map device drive label to serial numbers.
> Any thoughts how I could do this based on the data I have saved in that
> forum post?

Please show current 'smartctl -x' output for all of these devices, too.
 Just paste it all in your reply (with word wrap turned off).

Phil

[1] https://github.com/pturmel/lsdrv

  reply	other threads:[~2015-03-28  1:59 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <54ABEE54.6020707@sympatico.ca>
2015-01-07 13:34 ` mdadm RAID6 "active" with spares and failed disks; need help Matt Callaghan
2015-01-11 20:26   ` Matt Callaghan
2015-01-11 23:22     ` Valentijn Sessink
2015-01-12 16:35       ` Wols Lists
2015-01-21  0:34         ` Matt Callaghan
2015-01-22  9:47           ` Valentijn
2015-03-27 23:48             ` Matt Callaghan
2015-03-28  1:59               ` Phil Turmel [this message]
2015-03-28 10:11                 ` Roman Mamedov
2015-03-28 15:11                   ` Read-only mounts (was mdadm RAID6 "active" with spares and failed disks; need help) Phil Turmel
     [not found]                 ` <55161943.1090206@sympatico.ca>
     [not found]                   ` <BLU436-SMTP135A84B36120ACD11B2783D81F70@phx.gbl>
2015-03-28 17:40                     ` mdadm RAID6 "active" with spares and failed disks; need help Phil Turmel
2015-01-06 14:16 Matt Callaghan

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=55160B0A.1090405@turmel.org \
    --to=philip@turmel.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=matt_callaghan@sympatico.ca \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.