All of lore.kernel.org
 help / color / mirror / Atom feed
From: Wols Lists <antlists@youngman.org.uk>
To: Thomas Grawert <thomasgrawert0282@gmail.com>, linux-raid@vger.kernel.org
Cc: Phil Turmel <philip@turmel.org>
Subject: Re: help requested for mdadm grow error
Date: Mon, 25 May 2020 20:20:38 +0100	[thread overview]
Message-ID: <5ECC1A86.3010104@youngman.org.uk> (raw)
In-Reply-To: <4891e1e8-aaee-b36b-4131-ca7deb34defd@gmail.com>

On 25/05/20 20:05, Thomas Grawert wrote:
> The EFAX had me worried a moment, but these are 12TB Reds? That's fine.
>> A lot of the smaller drives are now shingled, ie not fit for purpose!
>>
>> Debian 10 - I don't know my Debians - how up to date is that? Is it a
>> new kernel with not much backports, or an old kernel full of backports?
>>
>> What version of mdadm?
>>
>>
>> That said, everything looks good. There are known problems - WITH FIXES
>> - growing a raid 5 so I suspect you've fallen foul of one. I'd sort out
>> a rescue disk that you can boot off as you might need it. Once we know a
>> bit more the fix is almost certainly a rescue disk and resume the
>> reshape, or a revert-reshape and then reshaping from a rescue disk. At
>> which point, you'll get your array back with everything intact.
> 
> yes, that´s the 12TB WD-Red - I´m using five pieces of it.
> 
> The Debian 10 is the most recent one. Kernel version is 4.9.0-12-amd64.
> mdadm-version is v3.4 from 28th Jan 2016 - seems to be the latest,
> because I can´t upgrade to any newer one using apt upgrade.

OW! OW! OW!

The newest mdadm is 4.1 or 4.2. UPGRADE NOW. Just download and build it
from the master repository - the instructions are in the wiki.

And if Debian 10 is the latest, kernel 4.9 will be a
franken-patched-to-hell-kernel ... I believe the latest kernel is 5.6?
> 
> I don´t think I need a rescue disk, because the raid isn´t bootable.
> It´s simply a big storage.
> 
Okay, the latest mdadm *might* fix your problem. However, you probably
need a proper up-to-date kernel as well, so you DO need a rescue disk.
Unless Debian has the option of upgrading the kernel to a 5.x series
kernel which hopefully isn't patched to hell and back?

This looks like the classic "I'm running ubuntu with a franken-kernel
and raid administration no longer works" problem.

I'm guessing (an educated "probably right" guess) that your reshape has
hung at 0% complete. So the fix is to get your rescue disk, use the
latest mdadm to do a revert-reshape, then use the latest kernel and
mdadm to do the reshape, before booting back in to your old Debian and
carrying as if nothing had happened.

Cheers,
Wol

  reply	other threads:[~2020-05-25 19:20 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-25 17:25 help requested for mdadm grow error Thomas Grawert
2020-05-25 18:09 ` Wols Lists
2020-05-25 18:18   ` Thomas Grawert
2020-05-25 18:55     ` Wols Lists
2020-05-25 19:05       ` Thomas Grawert
2020-05-25 19:20         ` Wols Lists [this message]
2020-05-25 21:45           ` John Stoffel
2020-05-25 21:55             ` antlists
2020-05-25 19:33         ` Mikael Abrahamsson
2020-05-25 19:35           ` Thomas Grawert
2020-05-25 19:42             ` Andy Smith
2020-05-25 20:30           ` Thomas Grawert
2020-05-25 21:19             ` antlists
2020-05-25 21:22               ` Thomas Grawert
2020-05-25 22:02                 ` antlists
2020-05-25 22:18                   ` Thomas Grawert
2020-05-25 22:32                     ` antlists
2020-05-25 23:01                       ` Thomas Grawert
2020-05-25 23:15                       ` Thomas Grawert
2020-05-25 23:31                       ` Thomas Grawert
2020-05-26  1:16                         ` antlists
2020-05-26  1:16                       ` Thomas Grawert
2020-05-25 18:24   ` Thomas Grawert

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=5ECC1A86.3010104@youngman.org.uk \
    --to=antlists@youngman.org.uk \
    --cc=linux-raid@vger.kernel.org \
    --cc=philip@turmel.org \
    --cc=thomasgrawert0282@gmail.com \
    /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.