[Tfug] Really slow mdadm resyncing
Matt Jacob
matt.jacob at gmail.com
Sun Jun 17 23:27:19 MST 2007
Hey guys:
I just set up a box with Debian etch, and I'm using mdadm with 2 IDE
drives in RAID1. I rsync'd about 50G of data to the box today, and
that finished at around 9:00 or 10:00 this morning. More than twelve
hours later, the RAID is still "active, resyncing", and I can't figure
out why. I don't even know what that means (I thought the status was
supposed to be "dirty" while recovering).
Here's the output from mdadm:
ned:~# mdadm -D /dev/md2
/dev/md2:
Version : 00.90.03
Creation Time : Sat Jun 16 11:45:51 2007
Raid Level : raid1
Array Size : 159139776 (151.77 GiB 162.96 GB)
Device Size : 159139776 (151.77 GiB 162.96 GB)
Raid Devices : 2
Total Devices : 2
Preferred Minor : 2
Persistence : Superblock is persistent
Update Time : Sat Jun 16 18:36:02 2007
State : active, resyncing
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Rebuild Status : 64% complete
UUID : 7f38bf9f:27f5bf50:05d2898a:691e38c1
Events : 0.9
Number Major Minor RaidDevice State
0 33 3 0 active sync /dev/hde3
1 33 67 1 active sync /dev/hdf3
And then there's this, which shows a pathetic 1M/s transfer rate:
ned:~# cat /proc/mdstat
Personalities : [raid1]
md2 : active raid1 hde3[0] hdf3[1]
159139776 blocks [2/2] [UU]
[============>........] resync = 64.3% (102406080/159139776)
finish=830.1min speed=1137K/sec
md1 : active raid1 hde2[0] hdf2[1]
746944 blocks [2/2] [UU]
md0 : active raid1 hde1[0] hdf1[1]
192640 blocks [2/2] [UU]
unused devices: <none>
The strange thing is that my speed_limit_min is set to 50000, which
should be 50M/s:
ned:~# cat /proc/sys/dev/raid/speed_limit_min
50000
Any ideas? Am I missing something? If I can't get better performance
out of this thing, I'm gonna go back to my poor man's RAID which
consisted of a nightly rsync job between the two drives.
Matt
P.S. Here are my system specs: Athlon 1200 MHz, 768 M RAM, (2) IDE ATA100 drives
More information about the tfug
mailing list