New raid5 took 5 days for initial build on spare drive

New raid5 took 5 days for initial build on spare drive

[Spencer Tuttle]

Ok, I just created my first ever raid 5 using software raid.

I am using a 2.6.14 kernel and the just newly supported MV88SX6041
Marvell SATA Controller. I received the following errors when the device
is initialized in the kernel

blk_queue_max_hw_segments: set to minimum 1

repeated a bunch of times.

And I wonder if this is why it is so slow, my /proc/mdstat shows the
following:
orion ~ # cat /proc/mdstat
Personalities : [raid5]
md_d0 : active raid5 sda1[0] sdd1[4] sdc1[2] sdb1[1]
      234444288 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
      [===================>.]  recovery = 98.9% (77292696/78148096)
      finish=102.9min speed=136K/sec

unused devices: <none>

As you can see it is almost finished, but it has been a little over 5
days, seems a little high, for 80GB drives, but then again this is my
first time using mdadm. 

Could someone let me know if this is normal or if I have something
really messed up.

Spencer

-- 
http://www.fastmail.fm - mmm... Fastmail...

Re: New raid5 took 5 days for initial build on spare drive

[Spencer Tuttle]

berk walker wrote:
> Just a couple of things to think about before the board explodes. :-)
> 
> Are your drives doing DMA?  Important.

Unfortunately no, these are serial ata drives and I saw this in the 
dmesg output:

libata version 1.12 loaded.
sata_mv version 0.12
ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 27 (level, low) -> IRQ 18
ata1: SATA max PIO4 cmd 0x0 ctl 0xF88A2120 bmdma 0x0 irq 18
ata2: SATA max PIO4 cmd 0x0 ctl 0xF88A4120 bmdma 0x0 irq 18
ata3: SATA max PIO4 cmd 0x0 ctl 0xF88A6120 bmdma 0x0 irq 18
ata4: SATA max PIO4 cmd 0x0 ctl 0xF88A8120 bmdma 0x0 irq 18
ata1: dev 0 cfg 49:2f00 82:7869 83:7d09 84:4003 85:7869 86:3c01 87:4003 
88:203f
ata1: dev 0 ATA, max UDMA/100, 156301488 sectors: lba48
ata1: dev 0 configured for PIO4
scsi0 : sata_mv
ata2: dev 0 cfg 49:2f00 82:7869 83:7d09 84:4003 85:7869 86:3c01 87:4003 
88:203f
ata2: dev 0 ATA, max UDMA/100, 156301488 sectors: lba48
ata2: dev 0 configured for PIO4
scsi1 : sata_mv
ata3: dev 0 cfg 49:2f00 82:7869 83:7d09 84:4003 85:7869 86:3c01 87:4003 
88:203f
ata3: dev 0 ATA, max UDMA/100, 156301488 sectors: lba48
ata3: dev 0 configured for PIO4
scsi2 : sata_mv
ata4: dev 0 cfg 49:2f00 82:7869 83:7d09 84:4003 85:7869 86:3c01 87:4003 
88:203f
ata4: dev 0 ATA, max UDMA/100, 156301488 sectors: lba48
ata4: dev 0 configured for PIO4
scsi3 : sata_mv
blk_queue_max_hw_segments: set to minimum 1
   Vendor: ATA       Model: Maxtor 6Y080M0    Rev: YAR5
   Type:   Direct-Access                      ANSI SCSI revision: 05
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1
blk_queue_max_hw_segments: set to minimum 1


I think that I am in PI04 mode, but this is an issue with the driver.

> 
> Are you using mdadm, or the compubox's firmware... (?)

The firmware for the controller is set to JBOD.
> 
> when you are (re)syncing, you can echo a min speed command, also.

I set the min to 1000

orion raid # cat /proc/sys/dev/raid/speed_limit_min
1000

It seems to me this might be more of a driver issue, can anyone elaborate?

Spencer

Re: New raid5 took 5 days for initial build on spare drive

[Spencer Tuttle]

Spencer Tuttle wrote:
> berk walker wrote:
> 
>> Just a couple of things to think about before the board explodes. :-)
>>
>> Are your drives doing DMA?  Important.

No, I was not using DMA. I found the patches for the driver I needed 
applied them now, I am creating an array on the same disks in 30 mins 
not 5 days.

Thank You,
Spencer