Re: Raid 5 - not clean and then a failure.
From: Robin Hill <hidden>
Date: 2009-08-25 09:34:37
On Tue Aug 25, 2009 at 01:40:31AM -0700, Jon Hardcastle wrote:
quoted hunk ↗ jump to hunk
--- On Tue, 25/8/09, Robin Hill <robin@robinhill.me.uk> wrote:quoted
From: Robin Hill <redacted> Subject: Re: Raid 5 - not clean and then a failure. To: linux-raid@vger.kernel.org Date: Tuesday, 25 August, 2009, 9:16 AM On Tue Aug 25, 2009 at 12:54:49AM -0700, Jon Hardcastle wrote:quoted
Guys, I have been having some problems with my arrays that I think i have nailed down to a pci controller (well I say that - it is always the drives connected to *a* controller but I have tried 2!) anyway the latest saga is i was trying some new kernel options last night - which didn't work.Did they have the same chipset? I had problems with PCI controllers on one of my systems, which turned out to be some sort of conflict between the onboard chipset and the chipset on the controllers. I found a PCI card with a different chipset and have had no issues since.They are/were cheapy little via ones from 'aria' I got a new one and installed it along side a week ago 1 drive on it to 'test', when my array came to do a scrub a week later I got a whole host of issues. I am not sure what the cause was but now either of the controllers seem work reliably. I have a pci express controller but my kernel doesnt (yet!) support pci express. Do you know of you can get sata 3 on pci? or is it too slow?
By SATA 3, I assume you're actually referring to SATA 3GBit/s (as SATA 3
has only just been ratified, and I doubt you can get it at all yet)?
You'll probably be able to find PCI cards that support it, but the
standard PCI bus (32-bit, 33MHz) only has a bandwidth of 1 GBit/s, so
can't even keep up with 1.5GBit/s SATA, let alone 3GBit/s. A 64-bit or
66MHz bus would do better though.
Cheers,
Robin
--
___
( ' } | Robin Hill [off-list ref] |
/ / ) | Little Jim says .... |
// !! | "He fallen in de water !!" | Attachments
- (unnamed) [application/pgp-signature] 198 bytes