Subject: md hangs while rebuilding

Hello All,

I have 9 disks raid 1. I pulled out 4 disks, and using raidhotadd I
triggered a rebuild on 3 of them. While rebuilding md1, the rebuilding
process is stuck at 0.0%. Below is a snapshot of "/proc/mdstat".

-----
md1 : active raid1 sdi2[12] sdh2[11] sdg2[10] sde2[4] sdd2[1] sdc2[0]
sdb2[2] sda2[5]
405504 blocks [9/5] [UUU_UU___]
[>....................] recovery = 0.0% (384/405504)
finish=176649.2min speed=0K/sec
-----

The finish="***" time is increasing constantly.

(1) What may be the cause. I have experienced it several times. There is
no heavy IO going-on on any of the partitions. Machine is kind of idle.
(2) Can we somehow stop the rebuilding process and restart it again?
(3) Rebooting will fix it. But I am trying to find a better solution.

Any help is highly appreciated.

Thanking You
Shesha

--
.-----.
/ \
{ o | o }
|
\_/



2004-10-01 21:19:27

by bert hubert

[permalink] [raw]
Subject: Re: md hangs while rebuilding

On Fri, Oct 01, 2004 at 12:16:51PM -0700, Shesha B. Sreenivasamurthy wrote:

> I have 9 disks raid 1. I pulled out 4 disks, and using raidhotadd I
> triggered a rebuild on 3 of them. While rebuilding md1, the rebuilding
> process is stuck at 0.0%. Below is a snapshot of "/proc/mdstat".

Please please please tell people what kernel you run with and your exact
configuration.

--
http://www.PowerDNS.com Open source, database driven DNS Software
http://lartc.org Linux Advanced Routing & Traffic Control HOWTO

Subject: Re: md hangs while rebuilding

I am sorry. :(

Here are the configurations ....

(1) The linux kernel version : 2.4.26


(2) md information
---------------------

/dev/md0 on / type ext3 (rw,data=journal)
none on /proc type proc (rw)
/dev/md2 on /var type ext3 (rw,data=journal)
/dev/md7 on /var/nas type ext3 (rw,data=journal)
/dev/md5 on /tmp type ext3 (rw,data=journal)
/dev/md6 on /var/log type ext3 (rw,data=journal)
/dev/md1 on /alt type ext3 (rw,data=journal)
/dev/md3 on /opt type ext3 (rw,data=journal)
/dev/md4 on /mnt type ext3 (rw,data=journal)


(3) Partition Information
----------------------------

All 9 disk (sda-sdi) is partitioned as follows

Disk /dev/sda: 250.0 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/sda1 1 51 409657 fd Linux raid autodetect
/dev/sda2 52 102 409657+ fd Linux raid autodetect
/dev/sda3 103 111 72292+ 82 Linux swap
/dev/sda4 112 30401 243304425 5 Extended
/dev/sda5 112 137 208844+ fd Linux raid autodetect
/dev/sda6 138 176 313267 fd Linux raid autodetect
/dev/sda7 177 179 24097 fd Linux raid autodetect
/dev/sda8 180 205 208844+ fd Linux raid autodetect
/dev/sda9 206 244 313267 fd Linux raid autodetect
/dev/sda10 245 267 184747 fd Linux raid autodetect
/dev/sda11 268 30401 242051352+ 83 Linux

---------------

Please let me know if any more info is required. Any help is regarded.

-Shesha

On Fri, 2004-10-01 at 13:57, bert hubert wrote:
> On Fri, Oct 01, 2004 at 12:16:51PM -0700, Shesha B. Sreenivasamurthy wrote:
>
> > I have 9 disks raid 1. I pulled out 4 disks, and using raidhotadd I
> > triggered a rebuild on 3 of them. While rebuilding md1, the rebuilding
> > process is stuck at 0.0%. Below is a snapshot of "/proc/mdstat".
>
> Please please please tell people what kernel you run with and your exact
> configuration.
--
.-----.
/ \
{ o | o }
|
\_/