RAID SYNC

RAID SYNC

[Mikael Chambon]

Hi Again,

My second question is about RAID sync.

Just for my curiosity, I am testing RAID1 on two empty 110 gig partitions, 
the system takes about 3h30 to resync, why is it taking so long ?

Here is my idea: As the RAID layer has nothing to do with the FS layer,
the raid driver (I use md) does not and can't see the difference between 
a used block of data and an unused block of data. That's why it has to 
resync everything. Am I right ?
 
Thanks guys,
--
Mikael Chambon

Re: RAID SYNC

[Avijit Pathania]

Looks like you do not have DMA enabled. `dmesg |grep -i dma` to see if 
system recognized it and `hdparm -d /dev/hda` to check if it is enabled.
Check if your IDE chipset is configured in the kernel. Or usr 
/etc/sysconfig/harddisks to enable it using hdparm.

It should take around 30mins for it to finish resync.

HTH
Avijit

Mikael Chambon wrote:

>Hi Again,
>
>My second question is about RAID sync.
>
>Just for my curiosity, I am testing RAID1 on two empty 110 gig partitions, 
>the system takes about 3h30 to resync, why is it taking so long ?
>
>Here is my idea: As the RAID layer has nothing to do with the FS layer,
>the raid driver (I use md) does not and can't see the difference between 
>a used block of data and an unused block of data. That's why it has to 
>resync everything. Am I right ?
> 
>Thanks guys,
>--
>Mikael Chambon
>
>
>-
>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
>
>
>  
>

Re: RAID sync

[Neil Brown]

On Monday October 1, kris@spylog.com wrote:
> 
>         Kernel 2.4.6.SuSE-4GB-SMP, 2 CPU, 2Gb RAM, 4 HDD SCSI, M/B Intel L440GX.
> Messages from dmesg:
snip
> md: now!
> md: sdb2's event counter: 0000001c
> md: sda2's event counter: 0000001d
snip
> 
>         Why RAID do not start synchronization ? It is normal ?

Yes.
A difference of 1 in the event counters isn't considered enough to
treat on of them as old, and presumably the newest one (sda2) was
marked clean.
This could happen if the array was shut down cleanly, the new super
block (with the dirty bit cleared) was written to sda2, but the new
superblock was NOT written to sdb2 for some reason.  In this situation
there is no need to resync the array.

Could this be what happened in your case?

NeilBrown


> 
> --
> Oleg A. Yurlov aka Kris Werewolf, SysAdmin      OAY100-RIPN
> mailto:kris@spylog.com                          +7 095 332-03-88
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: RAID sync

[Jakob Østergaard]

On Mon, Oct 01, 2001 at 06:27:53PM +0400, Oleg A. Yurlov wrote:
> 
>         Privet :-)
> 
>         Kernel 2.4.6.SuSE-4GB-SMP, 2 CPU, 2Gb RAM, 4 HDD SCSI, M/B Intel L440GX.
> Messages from dmesg:
> 
...
> md: sdc2 [events: 0000001e](write) sdc2's sb offset: 15815872
> md: considering sdb2 ...
> md:  adding sdb2 ...
> md:  adding sda2 ...
> md: created md0
> md: bind<sda2,1>
> md: bind<sdb2,2>
> md: running: <sdb2><sda2>
> md: now!
> md: sdb2's event counter: 0000001c
> md: sda2's event counter: 0000001d
> md: superblock update time inconsistency -- using the most recent one
> md: freshest: sda2
> md0: max total readahead window set to 508k
> md0: 1 data-disks, max readahead per data-disk: 508k
> raid1: device sdb2 operational as mirror 1
> raid1: device sda2 operational as mirror 0
> raid1: raid set md0 active with 2 out of 2 mirrors
> md: updating md0 RAID superblock on device
> md: sdb2 [events: 0000001e](write) sdb2's sb offset: 15815872
> md: sda2 [events: 0000001e](write) sda2's sb offset: 15815872
> md: ... autorun DONE.
> 
>         Why RAID do not start synchronization ? It is normal ?

Doesn't it ?

Try "cat /proc/mdstat"

Synchronization is a background operation - your array is functional
immediately.

(this behaviour was changed from the really really old RAID code in unpatched
 2.2 to standard 2.4)

-- 
................................................................
:   jakob@unthought.net   : And I see the elder races,         :
:.........................: putrid forms of man                :
:   Jakob Østergaard      : See him rise and claim the earth,  :
:        OZ9ABN           : his downfall is at hand.           :
:.........................:............{Konkhra}...............:

RAID sync

[Oleg A. Yurlov]

        Привет :-)

        Kernel 2.4.6.SuSE-4GB-SMP, 2 CPU, 2Gb RAM, 4 HDD SCSI, M/B Intel L440GX.
Messages from dmesg:

(scsi0:0:3:0) Synchronous at 80.0 Mbyte/sec, offset 63.
SCSI device sdd: 35843670 512-byte hdwr sectors (18352 MB)
 sdd: sdd1 sdd2
md: raid1 personality registered
md: raid5 personality registered
raid5: measuring checksumming speed
   8regs     :  1321.600 MB/sec
   32regs    :   978.400 MB/sec
   pIII_sse  :  1632.400 MB/sec
   pII_mmx   :  1790.000 MB/sec
   p5_mmx    :  1885.200 MB/sec
raid5: using function: pIII_sse (1632.400 MB/sec)
md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
md: Autodetecting RAID arrays.
(read) sda2's sb offset: 15815872 [events: 0000001d]
(read) sdb2's sb offset: 15815872 [events: 0000001c]
(read) sdc2's sb offset: 15815872 [events: 0000001d]
(read) sdd2's sb offset: 15815872 [events: 0000001d]
md: autorun ...
md: considering sdd2 ...
md:  adding sdd2 ...
md:  adding sdc2 ...
md: created md1
md: bind<sdc2,1>
md: bind<sdd2,2>
md: running: <sdd2><sdc2>
md: now!
md: sdd2's event counter: 0000001d
md: sdc2's event counter: 0000001d
md1: max total readahead window set to 508k
md1: 1 data-disks, max readahead per data-disk: 508k
raid1: device sdd2 operational as mirror 1
raid1: device sdc2 operational as mirror 0
raid1: raid set md1 active with 2 out of 2 mirrors
md: updating md1 RAID superblock on device
md: sdd2 [events: 0000001e](write) sdd2's sb offset: 15815872
md: sdc2 [events: 0000001e](write) sdc2's sb offset: 15815872
md: considering sdb2 ...
md:  adding sdb2 ...
md:  adding sda2 ...
md: created md0
md: bind<sda2,1>
md: bind<sdb2,2>
md: running: <sdb2><sda2>
md: now!
md: sdb2's event counter: 0000001c
md: sda2's event counter: 0000001d
md: superblock update time inconsistency -- using the most recent one
md: freshest: sda2
md0: max total readahead window set to 508k
md0: 1 data-disks, max readahead per data-disk: 508k
raid1: device sdb2 operational as mirror 1
raid1: device sda2 operational as mirror 0
raid1: raid set md0 active with 2 out of 2 mirrors
md: updating md0 RAID superblock on device
md: sdb2 [events: 0000001e](write) sdb2's sb offset: 15815872
md: sda2 [events: 0000001e](write) sda2's sb offset: 15815872
md: ... autorun DONE.

        Why RAID do not start synchronization ? It is normal ?

--
Oleg A. Yurlov aka Kris Werewolf, SysAdmin      OAY100-RIPN
mailto:kris@spylog.com                          +7 095 332-03-88