Jeremy Rosengren <jeremy.rosengren@...> writes:

Google:

http://www.google.com/search?hl=en&rlz=1B3GGGL_enUS326US326&q=mdadm+... j

/proc/sys/dev/raid/speed_limit_min /proc/sys/dev/raid/speed_limit_max

[root@localhost ~]# cat /proc/sys/dev/raid/speed_limit_min 10000 [root@localhost ~]# cat /proc/sys/dev/raid/speed_limit_max 1000000

Thanks

Rajagopal

This list never ceases to amaze me with very very quick and educating responses

John R Pierce <pierce@...> writes:

Rajagopal Swaminathan wrote:

depends on your disk controllers. whats `iostat -x 5` say the IO on hdb3 and hda3 is doing? (ignore the first sample, its average since

avg-cpu: %user %nice %system %iowait %steal %idle 0.13 0.00 31.98 0.00 0.00 67.89

Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util hda 26.40 0.40 4.20 2.20 3660.80 20.80 575.25 2.74 386.97 148.44 95.00 hdb 0.00 24.80 0.00 6.60 0.00 3988.80 604.36 1.43 217.45 143.24 94.54 md2 0.00 0.00 0.00 2.60 0.00 20.80 8.00 0.00 0.00 0.00 0.00 md1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 md0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 2.60 0.00 20.80 8.00 0.43 163.85 78.08 20.30

I won't claim I understand everyone of the above columns. I am in the process of reading man pages.

is the md2 filesystem mounted ?

Yes. But no users using the system and it is mounted as / (root). So cant unmount it

500gb has to be read and written over the SATA channels, which appear to be in IDE mode (hda instead of sda).... figure 60-80MB/sec peak sustainable, lets round that to 50MB/sec, so thats 500000/50 == 10000 seconds if its running at pure wire speed the whole time. if there's -any- disk contention, it will run MUCH slower. if its only running 2MB/sec (as indicated) somethinig is fubar with the disk IO, maybe its in PIO mode, ouch!

So

Q1. Where should I look to correct its PIO mode hdparm? BIOS? Q2. If I bring down the system, will the array reconstruction start from beginning or from where it left off before reboot?

BTW it is HP ML 110 G5 box with 4 gigs RAM.

Thanks

Rajagopal

A critical question repeated as I can't have this going on over the weekend and have to get the system up before next 18 hours

Q2. If I bring down the system, will the array reconstruction start from beginning or from where it left off before reboot?

Thanks

Rajagopal

John R Pierce wrote:

Rajagopal Swaminathan wrote:

...

A critical question repeated as I can't have this going on over the weekend and have to get the system up before next 18 hours

...

Q2. If I bring down the system, will the array reconstruction start from beginning or from where it left off before reboot?

it resumes from where it left off. it keeps track of the checkpoint locations via the metadata it stores in the last track of the partition (I think thats where it is)

It should only take a couple of hours anyway unless there is a lot of other activity on the partition. Something must be wrong with the controller or drive.

John R Pierce <pierce@...> writes:

Les Mikesell wrote:

...

It should only take a couple of hours anyway unless there is a lot of other activity on the partition. Something must be wrong with the controller or drive.

The IOSTAT output earlier showing the drives 100% busy at 1.8MB/sec tells me they are running in ISA PIO mode. there's a issue with the disk controller support and how its configured.

[SOLVED]

Thanks all

I powered down the box and changed the SATA mode in BIOS f4rom Auto to Serial ATA. That did the trick

cat /proc/mdstat Personalities : [raid1] md0 : active raid1 sdb1[1] sda1[0] 104320 blocks [2/2] [UU] resync=DELAYED

md1 : active raid1 sdb2[1] sda2[0] 4096448 blocks [2/2] [UU] resync=DELAYED

md2 : active raid1 sdb3[1] sda3[0] 484182912 blocks [2/2] [UU] [=>...................] resync = 5.6% (27346752/484182912) finish=68.6min speed=110970K/sec

unused devices: <none>

======= iostat -x 5 outpur (4th or 5th one)

avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 2.60 0.00 0.00 97.40

Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util sda 1752.89 0.00 259.68 0.00 257584.03 0.00 991.93 2.47 9.52 3.85 99.86 sda1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sda2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sda3 1752.89 0.00 259.68 0.00 257584.03 0.00 991.93 2.47 9.52 3.85 99.86 sdb 0.00 1752.69 0.00 259.68 0.00 257609.58 992.02 1.51 5.80 3.85 99.86 sdb1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sdb2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 sdb3 0.00 1752.69 0.00 259.68 0.00 257609.58 992.02 1.51 5.80 3.85 99.86 md2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 md1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 md0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 dm-0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

=========== hdparm -tT output

hdparm -tT /dev/sda

/dev/sda: Timing cached reads: 27444 MB in 1.99 seconds = 13802.85 MB/sec Timing buffered disk reads: 378 MB in 3.01 seconds = 125.67 MB/sec [root@localhost ~]# hdparm -tT /dev/sdb

/dev/sdb: Timing cached reads: 27116 MB in 1.99 seconds = 13636.74 MB/sec Timing buffered disk reads: 398 MB in 3.01 seconds = 132.35 MB/sec [root@localhost ~]#

Thanks again

Regards

Rajagopal

John R Pierce <pierce@...> writes:

Rajagopal Swaminathan wrote:

...

avg-cpu: %user %nice %system %iowait %steal %idle 0.00 0.00 2.60 0.00 0.00 97.40

and, note its now not only copying the data nearly 100X faster, its 97% idle. I believe your earlier iostat showed like 60% system time or something.

indeed

From earlier post..

avg-cpu: %user %nice %system %iowait %steal %idle 0.13 0.00 31.98 0.00 0.00 67.89

It through with syncing..

Thanks a lot again

Regards

Rajagopal