I have performance problems. I just downloaded and compiled
kernel 2.4.18 in the hopes that it would run better than the redhat
2.4.9-13 that caused data corruption on my raid system.
However when the only thing I have on the system is the normal
startup processes with NFS serving the array and then using tar
to recover data into the array from a remote system the load seems
to be used for non-raid activity.
Here is what most of the time takes the load
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
6 root 20 0 0 0 0 SW 42.1 0.0 1:56 kupdated
4 root 10 0 0 0 0 SW 13.5 0.0 1:58 kswapd
The lights on the drives hardly ever blink, data is moving
way too slow and I'm just getting frustrated as I don't see
any way in hell the 100GB will ever be recovered.
I know I must have forgotten something during the kernel
configuration or the array creation. Hopefully someone on
the list will know what I'm doing wrong.
I used mdadm to create the array (4 80GB Maxtor drives):
[root@fs01 etc]# mdadm --create /dev/md0 --level=0 --raid-disks=2
/dev/hdf1 /dev/hdh1
mdadm: array /dev/md0 started.
[root@fs01 etc]# mdadm --create /dev/md1 --level=0 --raid-disks=2
/dev/hdj1 /dev/hdl1
mdadm: array /dev/md1 started.
[root@fs01 etc]# mdadm --create /dev/md2 --level=1 --raid-disks=2
/dev/md0 /dev/md1
mdadm: array /dev/md2 started.
Here is the current output for mdadm detail info:
[root@fs01 etc]# mdadm -Q --detail /dev/md2
/dev/md2:
Version : 00.90.00
Creation Time : Fri Apr 5 22:32:43 2002
Raid Level : raid1
Array Size : 160071360 (152.65 GiB 163.91 GB)
Device Size : 160071360 (152.65 GiB 163.91 GB)
Raid Disks : 2
Total Disks : 2
Preferred Minor : 2
Persistance : Superblock is persistant
Update Time : Fri Apr 5 22:32:43 2002
State : dirty, no-errors
Active Drives : 2
Working Drives : 2
Failed Drives : 0
Spare Drives : 0
Number Major Minor RaidDisk State
0 9 0 0 active sync /dev/md0
1 9 1 1 active sync /dev/md1
UUID : 324ba547:7a880d5a:4034dc23:b95173a1
And here is the /proc/mdstat:
[root@fs01 etc]# cat /proc/mdstat
Personalities : [raid0] [raid1]
read_ahead 1024 sectors
md2 : active raid1 md1[1] md0[0]
160071360 blocks [2/2] [UU]
[>....................] resync = 0.2% (465000/160071360)
finish=19600.7min speed=133K/sec
md1 : active raid0 hdl1[1] hdj1[0]
160071424 blocks 64k chunks
md0 : active raid0 hdh1[1] hdf1[0]
160071424 blocks 64k chunks
unused devices: <none>
Another question I had is: How could I do md2 without
causing a resync? I think that the resync is a waste of
time as my next command is a mkreiserfs and I didn't
see it on the man page (hopefully I read it right)
Thanks in advance,
Alberto
-
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
