linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Xiao Ni <xni@redhat.com>
To: Chris Murphy <chris@colorremedies.com>
Cc: linux-raid@vger.kernel.org, eguan@redhat.com,
	Jes Sorensen <jes.sorensen@redhat.com>
Subject: Re: indefinite hang when growing/reshaping
Date: Fri, 19 Dec 2014 03:14:47 -0500 (EST)	[thread overview]
Message-ID: <997573837.677557.1418976887168.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <CAJCQCtQWSVbx08xafh6MDikiJOZJ6MVJ4Jx=qt=UZnX3t47A7A@mail.gmail.com>

Hi all

   There is the same problem. I can reproduce with the same steps. And
I can reproduce this in a different way too. I don't know whether it's 
the same problem, but the result is the same.

   The different steps are as follows:

mdadm -CR /dev/md0 -l5 -n7 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1 /dev/sdh1 -x1 /dev/sdi1  --assume-clean --bitmap=internal
mdadm --grow -l0 /dev/md0  --backup-file=tmp0
cat /proc/mdstat 
Personalities : [raid6] [raid5] [raid4] [raid1] [raid0] 
md0 : active raid5 loop3[4] loop2[3] loop1[1] loop0[0]
      2045952 blocks super 1.2 level 5, 512k chunk, algorithm 2 [4/4] [UUUU]
      [>....................]  reshape =  0.0% (0/1022976) finish=5.8min speed=2818K/sec
      
unused devices: <none>

   There is speed 2818K/sec, but if you look by iostat, there is no speed about the disks.


   And there is a workaround:
   mdadm -CR /dev/md0 -l5 -n7 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1 /dev/sdh1 -x1 /dev/sdi1  --assume-clean --bitmap=internal
   mdadm --grow -l0 /dev/md0

   If we don't specify the backup file. The reshape can finish correctly. 

Best Regards
Xiao

----- Original Message -----
> From: "Chris Murphy" <chris@colorremedies.com>
> To: linux-raid@vger.kernel.org
> Sent: Thursday, December 18, 2014 6:20:50 AM
> Subject: indefinite hang when growing/reshaping
> 
> If this is user error I'd expect a discreet error message, but I don't
> get one, just a hang during the reshape following successful --grow
> command (going from 3x device raid5 to 4x device raid5).
> 
> # cat /proc/mdstat
> Personalities : [raid6] [raid5] [raid4]
> md127 : active raid5 loop3[4] loop2[3] loop1[1] loop0[0]
>       2095104 blocks super 1.2 level 5, 512k chunk, algorithm 2 [4/4] [UUUU]
>       [>....................]  reshape =  0.0% (0/1047552)
> finish=7.0min speed=2402K/sec
> 
> Reproduces with:
> kernel-3.17.6-300.fc21.x86_64
> mdadm-3.3.2-1.fc21.x86_64
> and
> 3.18.0-2.fc22.i686+debug
> mdadm-3.3.2-1.fc21.i686
> 
> Details are here:
> https://bugzilla.kernel.org/show_bug.cgi?id=89851
> 
> --
> Chris Murphy
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

  reply	other threads:[~2014-12-19  8:14 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-17 22:20 indefinite hang when growing/reshaping Chris Murphy
2014-12-19  8:14 ` Xiao Ni [this message]
2014-12-19  8:17   ` Xiao Ni

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=997573837.677557.1418976887168.JavaMail.zimbra@redhat.com \
    --to=xni@redhat.com \
    --cc=chris@colorremedies.com \
    --cc=eguan@redhat.com \
    --cc=jes.sorensen@redhat.com \
    --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).