The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
I'm swapping out motherboards from an Nvidia Nforce 590 sli to a board based on the Intel P35 chipset. The board has the 2-port Jmicron SATA RAID controller, as well as the 4-port Intel ICH9 RAID controller.
My main question is this: Are either of these two controllers as flaky and fucked up as the Nvidia controller? I've had several RAID outages in the past that have forced me to use other methods of data redundancy, and I'd like to know if I can go back to using a mirroring RAID with the Intel or Jmicron chips without the RAID degrading and fucking up on me every week or so.
I'm swapping out motherboards from an Nvidia Nforce 590 sli to a board based on the Intel P35 chipset. The board has the 2-port Jmicron SATA RAID controller, as well as the 4-port Intel ICH9 RAID controller.
My main question is this: Are either of these two controllers as flaky and fucked up as the Nvidia controller? I've had several RAID outages in the past that have forced me to use other methods of data redundancy, and I'd like to know if I can go back to using a mirroring RAID with the Intel or Jmicron chips without the RAID degrading and fucking up on me every week or so.
I use The only time I've seen Intel RAID fuckup lately is when I hard-reset a system with an ICH7 controller. It usually results in 0-1 verification error.
Well, for right now the only thing that's on RAID is my games disk, which is a striping RAID that's going to have to be deleted when the new mobo comes. They're a pair of Hitachi Deskstars. The other disks that I had (HAD) in a mirroring RAID (the one that fucked up every week) are a pair of Samsungs. They're just regular-ass SATA disks, nothing too terribly fancy. I want to put the two Hitachi drives in RAID 0 again and I'd like to rebuild my mirroring RAID, if I won't have to rebuild it every week like I did on my old Nvidia controller
Well, for right now the only thing that's on RAID is my games disk, which is a striping RAID that's going to have to be deleted when the new mobo comes. They're a pair of Hitachi Deskstars. The other disks that I had (HAD) in a mirroring RAID (the one that fucked up every week) are a pair of Samsungs. They're just regular-ass SATA disks, nothing too terribly fancy. I want to put the two Hitachi drives in RAID 0 again and I'd like to rebuild my mirroring RAID, if I won't have to rebuild it every week like I did on my old Nvidia controller
I asked because I know certain models of Western Digital hard drives had some integrated error detection/correction in their firmware that caused them to be dropped by RAID controllers rediculously often.
I only trust Seagate drives in arrays now, but to be fair I haven't even touched a Samsung in years, so I wouldn't know about them.
So, would this Intel RAID be stable enough to, say, run Windows Vista 64 from a striped volume, or would I have to reinstall every week due to raid failures?
So, would this Intel RAID be stable enough to, say, run Windows Vista 64 from a striped volume, or would I have to reinstall every week due to raid failures?
Couldn't say, I only ever use it for RAID1, but the only time I've had errors was after Hard-Reboots during testing, and those were recoverable (as a RAID1).
I've been running RAID0 only for years. My current mobo is a P35 board (Gigabyte GA-P35-DS3R), I'm up to four months and counting with no need to reinstall. This is the with the Intel controller. Don't use the other controller, I tried it and my array didn't even last a week. My old P4 board went for two years with no need to recreate the RAID0 array using Intel's ICH5R chipset, so I have fairly strong confidence in their RAID0 support.
You may notice that the Intel AHCI controller takes a really long time to detect your hard drives. It does for me, anyway. It's never a problem, just really annoying when I'm waiting for the system to start from a cold boot. I've started putting it to sleep rather than shutting it down, so I can avoid the lengthy detection delay. This lead me to discover another issue, which is that sometimes I have no audio on resume, and I have to reboot in order to get it back. I think it's a RealTek driver issue, or something to do with the onboard RealTek audio chipset's "convenient" (note: sarcasm, it's totally not convenient) automatic detection. GigaByte support thinks it's because I'm "overclocking," asshats that they are. So I'm thinking I'll need an audio card to get away from this onboard audio junk. Those are the only two problems I've noticed with my P35 board anyway, and neither is really a showstopper.
Another question: is the Intel RAID controller flexible? With my old controller I could designate which ports were RAID and which were normal IDE style. If I have RAID enabled, do all disks on the controller have to be in a RAID? Can I have multiple RAIDs? (i.e., one 2-disk striping and one 2-disk mirroring, or 2 normal disks and a 2-disk mirroring RAID, etc.. etc..)
Posts
I use The only time I've seen Intel RAID fuckup lately is when I hard-reset a system with an ICH7 controller. It usually results in 0-1 verification error.
What kind of Hard drives are you using?
I asked because I know certain models of Western Digital hard drives had some integrated error detection/correction in their firmware that caused them to be dropped by RAID controllers rediculously often.
I only trust Seagate drives in arrays now, but to be fair I haven't even touched a Samsung in years, so I wouldn't know about them.
Couldn't say, I only ever use it for RAID1, but the only time I've had errors was after Hard-Reboots during testing, and those were recoverable (as a RAID1).
You may notice that the Intel AHCI controller takes a really long time to detect your hard drives. It does for me, anyway. It's never a problem, just really annoying when I'm waiting for the system to start from a cold boot. I've started putting it to sleep rather than shutting it down, so I can avoid the lengthy detection delay. This lead me to discover another issue, which is that sometimes I have no audio on resume, and I have to reboot in order to get it back. I think it's a RealTek driver issue, or something to do with the onboard RealTek audio chipset's "convenient" (note: sarcasm, it's totally not convenient) automatic detection. GigaByte support thinks it's because I'm "overclocking," asshats that they are. So I'm thinking I'll need an audio card to get away from this onboard audio junk. Those are the only two problems I've noticed with my P35 board anyway, and neither is really a showstopper.