From: Sebastian Reichel <elektranox@gmail.com>
To: Neil Brown <neilb@suse.de>
Cc: linux-raid@vger.kernel.org
Subject: Re: power outage while raid5->raid6 was in progress
Date: Wed, 14 Jul 2010 00:14:12 +0200 [thread overview]
Message-ID: <20100713221412.GA29708@earth.universe> (raw)
In-Reply-To: <20100709084808.72fc8fd4@notabene.brown>
[-- Attachment #1: Type: text/plain, Size: 4426 bytes --]
Hi,
I just did as you told:
* upgrading to 2.6.34.1
* go back to 5 disk raid5
* start raid5 -> raid6 again
This time I got no error and it began reshaping:
md0 : active raid6 sdk1[0] sdf1[5] sdg1[6] sdh1[7] sdi1[4] sdj1[3] sdl1[2] sde1[1]
5860543744 blocks super 0.91 level 6, 64k chunk, algorithm 18 [8/7] [UUUUU_UU]
[>....................] reshape = 2.8% (41047020/1465135936) finish=32916.5min speed=721K/sec
The reshaping got stuck again. Checking top I can see ksoftirqd uses 100% CPU on one
of the 2 cores. But there is one difference. This time there are no complains about any
disk device(s). Here is the dmesg output - The lower part (from 'blocked for more than
120 seconds') is repeated multiple times.
[ 612.736649] raid5: device sdk1 operational as raid disk 0
[ 612.736653] raid5: device sdi1 operational as raid disk 4
[ 612.736656] raid5: device sdj1 operational as raid disk 3
[ 612.736658] raid5: device sdl1 operational as raid disk 2
[ 612.736660] raid5: device sde1 operational as raid disk 1
[ 612.737301] raid5: allocated 6386kB for md0
[ 612.749007] 0: w=1 pa=0 pr=6 m=2 a=18 r=6 op1=0 op2=0
[ 612.749011] 4: w=2 pa=0 pr=6 m=2 a=18 r=6 op1=0 op2=0
[ 612.749014] 3: w=3 pa=0 pr=6 m=2 a=18 r=6 op1=0 op2=0
[ 612.749016] 2: w=4 pa=0 pr=6 m=2 a=18 r=6 op1=0 op2=0
[ 612.749018] 1: w=5 pa=0 pr=6 m=2 a=18 r=6 op1=0 op2=0
[ 612.749021] raid5: raid level 6 set md0 active with 5 out of 6 devices, algorithm 18
[ 612.749065] RAID5 conf printout:
[ 612.749067] --- rd:6 wd:5
[ 612.749069] disk 0, o:1, dev:sdk1
[ 612.749071] disk 1, o:1, dev:sde1
[ 612.749073] disk 2, o:1, dev:sdl1
[ 612.749075] disk 3, o:1, dev:sdj1
[ 612.749077] disk 4, o:1, dev:sdi1
[ 612.779095] RAID5 conf printout:
[ 612.779099] --- rd:8 wd:7
[ 612.779102] disk 0, o:1, dev:sdk1
[ 612.779104] disk 1, o:1, dev:sde1
[ 612.779105] disk 2, o:1, dev:sdl1
[ 612.779107] disk 3, o:1, dev:sdj1
[ 612.779109] disk 4, o:1, dev:sdi1
[ 612.779111] disk 5, o:1, dev:sdf1
[ 612.779117] RAID5 conf printout:
[ 612.779118] --- rd:8 wd:7
[ 612.779120] disk 0, o:1, dev:sdk1
[ 612.779122] disk 1, o:1, dev:sde1
[ 612.779123] disk 2, o:1, dev:sdl1
[ 612.779125] disk 3, o:1, dev:sdj1
[ 612.779127] disk 4, o:1, dev:sdi1
[ 612.779129] disk 5, o:1, dev:sdf1
[ 612.779130] disk 6, o:1, dev:sdg1
[ 612.779134] RAID5 conf printout:
[ 612.779135] --- rd:8 wd:7
[ 612.779136] disk 0, o:1, dev:sdk1
[ 612.779138] disk 1, o:1, dev:sde1
[ 612.779140] disk 2, o:1, dev:sdl1
[ 612.779142] disk 3, o:1, dev:sdj1
[ 612.779143] disk 4, o:1, dev:sdi1
[ 612.779145] disk 5, o:1, dev:sdf1
[ 612.779147] disk 6, o:1, dev:sdg1
[ 612.779149] disk 7, o:1, dev:sdh1
[ 612.779233] md: reshape of RAID array md0
[ 612.779235] md: minimum _guaranteed_ speed: 1000 KB/sec/disk.
[ 612.779237] md: using maximum available idle IO bandwidth (but not more than 200000 KB/sec) for reshape.
[ 612.779244] md: using 128k window, over a total of 1465135936 blocks.
[ 2846.784515] Clocksource tsc unstable (delta = 4398042075725 ns)
[ 3000.536656] INFO: task md0_reshape:5554 blocked for more than 120 seconds.
[ 3000.536675] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 3000.536699] md0_reshape D 0000000100099eb1 0 5554 2 0x00000000
[ 3000.536704] ffff88007e08f700 0000000000000046 0000000000000082 ffff88006d3d1290
[ 3000.536707] ffffffff81632020 0000000000015280 0000000000015280 0000000000015280
[ 3000.536711] ffff88007e1f3fd8 0000000000015280 ffff88007e08f700 ffff88007e1f3fd8
[ 3000.536714] Call Trace:
[ 3000.536735] [<ffffffffa019e735>] ? reshape_request+0x1e7/0x7a8 [raid456]
[ 3000.536741] [<ffffffff8105d519>] ? autoremove_wake_function+0x0/0x2a
[ 3000.536746] [<ffffffffa019ee87>] ? sync_request+0x191/0x2d1 [raid456]
[ 3000.536754] [<ffffffffa0180a0b>] ? is_mddev_idle+0xa2/0xf5 [md_mod]
[ 3000.536760] [<ffffffffa0181171>] ? md_do_sync+0x713/0xb0f [md_mod]
[ 3000.536764] [<ffffffff8103767e>] ? update_curr+0xa2/0x145
[ 3000.536771] [<ffffffffa0181d4a>] ? md_thread+0xf2/0x110 [md_mod]
[ 3000.536777] [<ffffffffa0181c58>] ? md_thread+0x0/0x110 [md_mod]
[ 3000.536779] [<ffffffff8105d0e5>] ? kthread+0x75/0x7d
[ 3000.536783] [<ffffffff810097e4>] ? kernel_thread_helper+0x4/0x10
[ 3000.536786] [<ffffffff8105d070>] ? kthread+0x0/0x7d
[ 3000.536788] [<ffffffff810097e0>] ? kernel_thread_helper+0x0/0x10
-- Sebastian
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]
prev parent reply other threads:[~2010-07-13 22:14 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-07 20:41 power outage while raid5->raid6 was in progress Sebastian Reichel
2010-07-07 22:44 ` Neil Brown
2010-07-07 23:13 ` Sebastian Reichel
2010-07-08 1:21 ` Neil Brown
2010-07-08 8:48 ` Sebastian Reichel
2010-07-08 22:48 ` Neil Brown
2010-07-08 23:50 ` Sebastian Reichel
2010-07-13 22:14 ` Sebastian Reichel [this message]
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=20100713221412.GA29708@earth.universe \
--to=elektranox@gmail.com \
--cc=linux-raid@vger.kernel.org \
--cc=neilb@suse.de \
/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