Re: Does --freeze-reshape only work during initrd boot phase?

[NeilBrown <neilb@suse.de>]

[-- Attachment #1: Type: text/plain, Size: 3584 bytes --]

On Fri, 22 May 2015 06:30:09 -0400 (EDT) Xiao Ni <xni@redhat.com> wrote:

> Hi all
> 
>    I'm trying to use --freeze-reshape, I search how to use it, there is just some 
> explanation in man mdadm. And I see the main usage about this is during initrd boot phase.
> Does --freeze-reshape only work during initrd boot phase?

No it can be used at any time.
However it doesn't apply to all array types, and that isn't clear from the
documentation.

--freeze-reshape is only relevant for arrays which require a user-space
process to help with the reshape - specifically to make temporary backups of
portions of the array while they are being reshaped.
If --freeze-reshape is given, then when mdadm discovers that it needs to
start such a helper, it doesn't and just leaves reshape frozen.
If it never finds that a helper is needed, there is no value in freezing the
reshape.

I should probably improve the documentation.

NeilBrown

> 
>    I tried this after the machine started. The steps are as follows:
> [root@intel-canoepass-02 tmp]# mdadm -CR /dev/md0 -l5 -n5 /dev/loop[0-4] --assume-clean
> mdadm: /dev/loop0 appears to be part of a raid array:
>        level=raid5 devices=5 ctime=Fri May 22 06:14:35 2015
> mdadm: /dev/loop1 appears to be part of a raid array:
>        level=raid5 devices=5 ctime=Fri May 22 06:14:35 2015
> mdadm: /dev/loop2 appears to be part of a raid array:
>        level=raid5 devices=5 ctime=Fri May 22 06:14:35 2015
> mdadm: /dev/loop3 appears to be part of a raid array:
>        level=raid5 devices=5 ctime=Fri May 22 06:14:35 2015
> mdadm: /dev/loop4 appears to be part of a raid array:
>        level=raid5 devices=5 ctime=Fri May 22 06:14:35 2015
> mdadm: Defaulting to version 1.2 metadata
> mdadm: array /dev/md0 started.
> [root@intel-canoepass-02 tmp]# mdadm /dev/md0 -a /dev/loop5 
> mdadm: added /dev/loop5
> [root@intel-canoepass-02 tmp]# mdadm -Es
> ARRAY /dev/md/0  metadata=1.2 UUID=24d3d007:8ec28d4e:edb482fe:db7cab71 name=0
>    spares=1
> [root@intel-canoepass-02 tmp]# mdadm -Es > /etc/mdadm.conf 
> [root@intel-canoepass-02 tmp]# mdadm --grow /dev/md0 --raid-devices 6
> [root@intel-canoepass-02 tmp]# cat /proc/mdstat 
> Personalities : [raid6] [raid5] [raid4] 
> md0 : active raid5 loop5[5] loop4[4] loop3[3] loop2[2] loop1[1] loop0[0]
>       2043904 blocks super 1.2 level 5, 512k chunk, algorithm 2 [6/6] [UUUUUU]
>       [====>................]  reshape = 23.9% (122368/510976) finish=0.8min speed=7648K/sec
>       
> unused devices: <none>
> [root@intel-canoepass-02 tmp]# mdadm -S /dev/md0 
> mdadm: stopped /dev/md0
> [root@intel-canoepass-02 tmp]# mdadm -As --freeze-reshape
> mdadm: /dev/md/0 has been started with 6 drives.
> [root@intel-canoepass-02 tmp]# cat /proc/mdstat 
> Personalities : [raid6] [raid5] [raid4] 
> md0 : active raid5 loop0[0] loop5[5] loop4[4] loop3[3] loop2[2] loop1[1]
>       2043904 blocks super 1.2 level 5, 512k chunk, algorithm 2 [6/6] [UUUUUU]
>       [=========>...........]  reshape = 46.9% (240128/510976) finish=0.1min speed=23808K/sec
>       
> unused devices: <none>
> 
>    The reshape is not freeze. I think it should be freeze now, am I right? The reason I want to
> use this is that I want to use --grow --continue. I see --continue is used when reshape is 
> interrupted. 
> 
> Best Regards
> Xiao
> --
> 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


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 811 bytes --]