Hi - We''re looking for a reliable Athlon 64 motherboard. Hopefully this will support 4600+ (dual core) processors, SATA, and hopefully gigabit networking. Reliability is the key. We''ve tried using an MSI K8NGM2-FID, based on the nforce chipset. This works fine and the gigabit forcedeth appears reliable with the latest kernel update. However it oopses every few days for no apparent reason. We''ve also tried an Asus A8V-E SE (VIA chipset), but this appears terrible. The sky2 driver works (with the latest kernel update for 4.2, or the jwl test kernels), but then hangs after an hour or two of transferring data. We then tried a tg3 network card and an e1000, and this corrupts the data transferred between it and another computer. Maybe it is the SATA corrupting the data, but who knows. No obvious errors in the logs except for lost ticks. Tried non-SMP, 32 bit, noapic, acpi=off. No luck. Does anyone have a recommendation for a reliable board? Otherwise we have to buy Intel and get the speed loss... Many thanks Jeremy -- Jeremy Sanders <jss@ast.cam.ac.uk> http://www-xray.ast.cam.ac.uk/~jss/ X-Ray Group, Institute of Astronomy, University of Cambridge, UK. Public Key Server PGP Key ID: E1AAE053
Jeremy Sanders wrote:> Hi - We''re looking for a reliable Athlon 64 motherboard. Hopefully > this will support 4600+ (dual core) processors, SATA, and hopefully > gigabit networking. Reliability is the key. > > We''ve tried using an MSI K8NGM2-FID, based on the nforce chipset. This > works fine and the gigabit forcedeth appears reliable with the latest > kernel update. However it oopses every few days for no apparent reason. > > We''ve also tried an Asus A8V-E SE (VIA chipset), but this appears > terrible. The sky2 driver works (with the latest kernel update for > 4.2, or the jwl test kernels), but then hangs after an hour or two of > transferring data. We then tried a tg3 network card and an e1000, and > this corrupts the data transferred between it and another computer. > Maybe it is the SATA corrupting the data, but who knows. No obvious > errors in the logs except for lost ticks. Tried non-SMP, 32 bit, > noapic, acpi=off. No luck.We have been happy with the Giga-byte K8N51GMF-9 and the ASUS A8N-VM CSM. A few things: -Have you run Memtest86 to verify that you don''t have bad RAM? Let it run overnight for a few passes to verify absolute stability -Have you run Prime95 to verify your CPU isn''t bad or overheating? Again, let it run overnight to verify your CPU is being cooled proper, also keep the case closed. -Have you made certain there''s not an issue with the Network switch? Jay -------------- next part -------------- A non-text attachment was scrubbed... Name: jlee.vcf Type: text/x-vcard Size: 193 bytes Desc: not available Url : http://lists.centos.org/pipermail/centos/attachments/20060321/e77e374a/jlee.vcf
On Tue, 21 Mar 2006, Jay Lee wrote:> We have been happy with the Giga-byte K8N51GMF-9 and the ASUS A8N-VM CSM. A > few things: > > -Have you run Memtest86 to verify that you don''t have bad RAM? Let it > run overnight for a few passes to verify absolute stabilityYes - no obvious problem on either. I ran a repeated mozilla build on the K8NGM2-FID, copying large files across the network, and it worked fine for days. It then oopsed when it wasn''t doing anything, twice!> -Have you run Prime95 to verify your CPU isn''t bad or overheating?No - good idea.> Again, let it run overnight to verify your CPU is being cooled proper, also > keep the case closed. > -Have you made certain there''s not an issue with the Network switch?Pretty sure. It failed across a managed switch and an unmanaged one. Other e1000/e100 computers work fine with both switches. The Asus A8V-E SE is extremely easy to make fail. There is even data corruption with an e100 card. Jeremy -- Jeremy Sanders <jss@ast.cam.ac.uk> http://www-xray.ast.cam.ac.uk/~jss/ X-Ray Group, Institute of Astronomy, University of Cambridge, UK. Public Key Server PGP Key ID: E1AAE053