I can confirm that the AOC-SAT2-MV8 PCI-X card does work in a 32 bit PCI slot, at least on the Tyan 2927G2NR motherboards. Downside is that bandwidth is limited to ~120MB.
Since using the AOC-SAT2-MV8 instead of the on-board Nvidia MCP-55 I have not had any problems with the disks and I now have stable machines. Thanks again for all suggestions. -- Dave On 02/06/2008 05:18 PM, Al Hopper wrote: > On Wed, 6 Feb 2008, Dave wrote: > >> Just looked into add-in cards to get around the issue. Unfortunately >> the Tyan 2927 boards have only PCI-E and 32 bit PCI slots. I can't >> find any PCI-E SATA HBAs that have the Marvell 88SX chips, either. > > Yep - the lack of a cost-effective PCI-E board is an issue currently > hurting widespread ZFS adoption/experimentation. If only Sun would come > to the rescuse (hint, hint). You could look at an LsiLogic SAS board > with a PCI-E connector. I've also heard that some users have found the > SuperMicro board to work in a 32-bit slot and others have found that it > won't. I have no clue as to why this is the case. > > The only other alternative I'm aware of is a 3124 based PCI card. Let > me check the solaris on x86 (yahoo) list and email you a recommendation. > I think Bob Palowoda found a card that works... (I've CCed him). > >> I'm extremely disappointed that the SXCE SATA driver won't work with >> my MCP55 based controller. Seems like there should be no reason why it >> wouldn't since some Sun servers use MCP55 as well. > > :( > >> -- >> Dave >> >> On 02/06/2008 04:49 PM, Al Hopper wrote: >>> On Wed, 6 Feb 2008, Dave wrote: >>> >>>> I've got a number of machines running nv_b81 and I'm getting messages >>>> like this: >>>> >>>> WARNING: /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/[EMAIL >>>> PROTECTED],0 (sd0): >>>> Error for Command: read(10) Error Level: >>>> Retryable >>>> Requested Block: 10089026 Error Block: >>>> 10089026 >>>> Vendor: ATA Serial Number: >>>> >>>> Sense Key: No_Additional_Sense >>>> ASC: 0x0 (no additional sense info), ASCQ: 0x0, FRU: 0x0 >>>> NOTICE: /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]: >>>> port 0: device reset >>>> >>>> I'm unable to use the machines after I get the warning - they must be >>>> hard rebooted. I've tried disabling NCQ in /etc/system without success. >>>> This has happened in prior versions of Nevada as well. I usually get >>>> the >>>> message when first booting, but not always. >>>> >>>> All machines are running Dual Proc Opteron 2214, Tyan 2927G2NR >>>> motherboard, WD 500GB SATA drives. >>>> >>>> I'm at my wits end trying to find the solution. Any help is >>>> appreciated. >>> >>> Would installing a SuperMicro AOC-SAT2-MV8 PCI-X resolve this issue >>> (for now)? >>> >>> Regards, >>> >>> Al Hopper Logical Approach Inc, Plano, TX. [EMAIL PROTECTED] >>> Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT >>> OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007 >>> http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/ >> > > Al Hopper Logical Approach Inc, Plano, TX. [EMAIL PROTECTED] > Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT > OpenSolaris Governing Board (OGB) Member - Apr 2005 to Mar 2007 > http://www.opensolaris.org/os/community/ogb/ogb_2005-2007/ _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
