Kent Watsen wrote: > > Thanks Richard and Al, > > I'll refrain from express how disturbing this is, as I'm trying to > help the Internet be kid-safe ;) > > As for the PSU, I'd be very surprised there if that were it as it is a > 3+1 redundant PSU that came with this system, built by a reputable > integrator. Also, the PSU is plugged into a high-end APC UPS, sucking > just 25% of its capacity. And the UPS has a dedicated 240V 30A circuit. > > As for the memory, it might be - even though the same integrator > installed the SIMMs and did a 24-hour burn-in test, you never know. > So I'm running memtest86 now, which is 12% passed so far... > > I'm going to try another hardware test, which is to switch around the > backplanes my cards are plugging into. If the same backplanes are > failing, then I know all my AOC-SAT2-MV8 cards are OK. Likewise, if > the same backplane doesn't fail, then I know all my backplanes are > OK. Either way, I'll eliminate one potential hardware issue. >
You could also try the SunVTS system tests. They are what we use in the factory to prove systems work before being shipped to customers. Located in /usr/sunvts where READMEs, man pages, and binaries live. There are a zillion options, but I highly recommend the readonly disk tests for your case. > But I still think that it might be software related. My first post > was trying to point out some anomalies in how the devices are being > named - see the highlighted parts below? - doesn't that look strange? > - why would Solaris use different naming convention for some disks? > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED]/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED],1/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED],1/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED],1/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > /[EMAIL PROTECTED],0/pci10de,[EMAIL PROTECTED]/pci1033,[EMAIL > PROTECTED],1/pci11ab,[EMAIL PROTECTED]/[EMAIL PROTECTED],0 > > > PS: in case you can't see it, look at the last four disks and > notice how the contain a spurious ",1" and also have the same > "@6" as the middle four disks > > looks reasonable to me. These are just PCI device identifiers, certainly nothing to be worried about. -- richard _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
