I'm wondering about the following messages, which appeared when I upgraded from
2.4 to 2.6:
| sda: asking for cache data failed
| sda: assuming drive cache: write through
(a larger log snippet below)
I have found another message:
http://groups.google.com/group/fa.linux.kernel/browse_thread/thread/853b2b5a5cc2f837/87bcabcf8a07f065
whosese author asks about whether this means that data inconsistencies can
appear. I also have a megaraid card, and a similar situation, but I have a
bit more general question.
I'm wondering WHY can't the caching mode cannot be obtained by the kernel?
- Is it because the kernel is trying to inquire about the cache on a physical
disk, and the sda and sdb it sees are actually logical drives, and the
situation is perfectly normal?
- Or is it something wrong with the card?
- Or maybe there is no support for such operation in the megaraid
driver?
- Is it something else even? Do I need to do something more to find
out?
The message is simply a little worrying for me, because I don't understand why
the "asking" fails.
here's some context:
| megaraid cmm: 2.20.2.5 (Release Date: Fri Jan 21 00:01:03 EST 2005)
| SCSI subsystem initialized
| megaraid: 2.20.4.5 (Release Date: Thu Feb 03 12:27:22 EST 2005)
| megaraid: probe new device 0x1000:0x0407:0x8086:0x0532: bus 5:slot 0:func 0
| ACPI: PCI Interrupt 0000:05:00.0[A] -> GSI 52 (level, low) -> IRQ 169
| megaraid: fw version:[411M] bios version:[H404]
| scsi0 : LSI Logic MegaRAID driver
| scsi[0]: scanning scsi channel 0 [Phy 0] for non-raid devices
| scsi[0]: scanning scsi channel 1 [Phy 1] for non-raid devices
| scsi[0]: scanning scsi channel 2 [virtual] for logical drives
| Vendor: MegaRAID Model: LD 0 RAID1 35G Rev: 411M
| Type: Direct-Access ANSI SCSI revision: 02
| Vendor: MegaRAID Model: LD 1 RAID5 245G Rev: 411M
| Type: Direct-Access ANSI SCSI revision: 02
[...]
| SCSI device sda: 71772160 512-byte hdwr sectors (36747 MB)
| sda: asking for cache data failed
| sda: assuming drive cache: write through
| SCSI device sda: 71772160 512-byte hdwr sectors (36747 MB)
| sda: asking for cache data failed
| sda: assuming drive cache: write through
| /dev/scsi/host0/bus2/target0/lun0: p1 p2 p3 p4 < p5 p6 p7 >
| Attached scsi disk sda at scsi0, channel 2, id 0, lun 0
| SCSI device sdb: 502720512 512-byte hdwr sectors (257393 MB)
| sdb: asking for cache data failed
| sdb: assuming drive cache: write through
| SCSI device sdb: 502720512 512-byte hdwr sectors (257393 MB)
| sdb: asking for cache data failed
| sdb: assuming drive cache: write through
| /dev/scsi/host0/bus2/target1/lun0: p1 p2
| Attached scsi disk sdb at scsi0, channel 2, id 1, lun 0
This is kernel 2.6.12 with Debian patches.
regards,
Marcin
--
Marcin Owsiany <[email protected]> http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
"Every program in development at MIT expands until it can read mail."
-- Unknown
On Iau, 2005-10-13 at 12:45 +0200, Marcin Owsiany wrote:
> I'm wondering about the following messages, which appeared when I upgraded from
> 2.4 to 2.6:
>
> | sda: asking for cache data failed
> | sda: assuming drive cache: write through
>
> (a larger log snippet below)
The kernel asks the SCSI drive for its cache parameters. The AMI raid
card sitting in the middle doesn't know how to handle this so this
message occurs. It should be ok providing the raid card itself is
handling the consistency correctly but check with your card vendor.
On Thu, Oct 13 2005, Alan Cox wrote:
> On Iau, 2005-10-13 at 12:45 +0200, Marcin Owsiany wrote:
> > I'm wondering about the following messages, which appeared when I upgraded from
> > 2.4 to 2.6:
> >
> > | sda: asking for cache data failed
> > | sda: assuming drive cache: write through
> >
> > (a larger log snippet below)
>
> The kernel asks the SCSI drive for its cache parameters. The AMI raid
> card sitting in the middle doesn't know how to handle this so this
> message occurs. It should be ok providing the raid card itself is
> handling the consistency correctly but check with your card vendor.
The assumption that sd makes might be a little on the risky side though,
would seem a lot safer to assume write back if you don't know the
policy. At least that wont bring surprises later on...
--
Jens Axboe
On Thu, Oct 13, 2005 at 01:55:54PM +0100, Alan Cox wrote:
> On Iau, 2005-10-13 at 12:45 +0200, Marcin Owsiany wrote:
> > I'm wondering about the following messages, which appeared when I upgraded from
> > 2.4 to 2.6:
> >
> > | sda: asking for cache data failed
> > | sda: assuming drive cache: write through
> >
> > (a larger log snippet below)
>
[...]
> It should be ok providing the raid card itself is
> handling the consistency correctly but check with your card vendor.
Does that depend on the card only, or rather on the driver?
(megaraid_mbox in this case) If the latter, then maybe someone familiar
with its code can tell if the driver is OK in this regard?
thanks,
Marcin
--
Marcin Owsiany <[email protected]> http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
"Every program in development at MIT expands until it can read mail."
-- Unknown