Is it possible to pause/abort mdadm reshape?

Question

Saturday I added a new drive to my mdadm array, and the reshape process is taking a long time. Unfortunately there's a nasty storm coming, and I'm likely going to lose power, which will probably do bad things to my mdadm grow operation.

Current Status

# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid5 sdh1[6] sdc1[0] sdg1[5] sdf1[3] sde1[2] sdd1[1]
    11720534016 blocks super 1.2 level 5, 512k chunk, algorithm 2 [6/6] [UUUUUU]
    [=====>...............]  reshape = 25.9% (759125024/2930133504) finish=4091.0min speed=8844K/sec

mdadm --detail /dev/md0

 mdadm --detail /dev/md0
/dev/md0:
        Version : 1.2
  Creation Time : Sat Mar 23 07:41:24 2013
     Raid Level : raid5
     Array Size : 11720534016 (11177.57 GiB 12001.83 GB)
  Used Dev Size : 2930133504 (2794.39 GiB 3000.46 GB)
   Raid Devices : 6
  Total Devices : 6
    Persistence : Superblock is persistent

    Update Time : Tue May 13 11:34:08 2014
          State : clean, reshaping
 Active Devices : 6
Working Devices : 6
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 512K

 Reshape Status : 26% complete
  Delta Devices : 1, (5->6)

           Name : foxery:0  (local to host foxery)
           UUID : afbbf397:afc6ac78:6e7d376b:eedc2dd0
         Events : 9641

    Number   Major   Minor   RaidDevice State
       0       8       33        0      active sync   /dev/sdc1
       1       8       49        1      active sync   /dev/sdd1
       2       8       65        2      active sync   /dev/sde1
       3       8       81        3      active sync   /dev/sdf1
       5       8       97        4      active sync   /dev/sdg1
       6       8      113        5      active sync   /dev/sdh1

1. Is it possible to stop/pause the mdadm grow process?
2. Can I stop/abort the reshape process, and run with bitmaps?

I used this doc to improve speeds but it didn't do all that much since most of the speed would come from enabling bitmapping, which isn't doable on the fly, afaik.

Solution 1

echo frozen > /sys/block/md0/md/sync_action

Idle will move the reshare to the next block , ex if it was md0 , it will reshape md1. If you try it again , it will back reshape md0.

So frozen will pause them with indication: resync=PENDING

Solution 2

You can try to suspend it by issuing:-

echo "idle" >  /sys/block/md0/md/sync_action

As root.

Then immediately shut it down, so nothing else kicks in a reshape.