I have found additional information that may be useful regarding the hard
hang I have been experiencing.  Details were present in a previous message
but here is a short summary.

I have two of the exact same boxes (even same HD).  They are i810 mobos
with a Celeron 500 and 128MB ram and a 10GB IDE hard drive.  One runs
-stable and one runs -current.  The -stable box has never crashed but the
-current one used to hang every night (power light off, numlock stuck
on).  I swapped memory, drives, etc. -- same result: only the -current box
would crash.  I disabled ACPI and the -current box would stay up for a
few days to a week before hanging.  Now I removed the IDE hard drive and
replaced it with a SCSI one and a Adaptec 2940 U2W HBA.  Not only is the
IO much faster but it has run for weeks without a hang.

I would appreciate it if someone could help me look into this
further.  Are there known problems with this ATA chipset?  It was running
at UDMA33.

* Swapping all hardware with the -stable box doesn't change anything
* Disabling ACPI helps it run ~1 week before hanging
* Swapping SCSI drive for IDE keeps it from hanging (at least for a few
weeks)

-Nate

Old dmesg: (w/ ATA drive)
Timecounter "i8254"  frequency 1193182 Hz
Timecounter "TSC"  frequency 498486405 Hz
CPU: Pentium II/Pentium II Xeon/Celeron (498.49-MHz 686-class CPU)
  Origin = "GenuineIntel"  Id = 0x665  Stepping = 5
real memory  = 133103616 (129984K bytes)
avail memory = 124231680 (121320K bytes)
pcib0: <Intel 82810 (i810 GMCH) Host To Hub bridge> at pcibus 0 on
motherboard
pci0: <PCI bus> on pcib0
atapci0: <Intel ICH0 ATA33 controller> port 0x1800-0x180f at device 31.1
on pci0
ata0: at 0x1f0 irq 14 on atapci0
ata1: at 0x170 irq 15 on atapci0
ad0: DMA limited to UDMA33, non-ATA66 cable or device
ad0: 8297MB <Maxtor 90841U2> [16858/16/63] at ata0-master UDMA33

atapci0@pci0:31:1:      class=0x010180 card=0x24118086 chip=0x24118086
rev=0x02 hdr=0x00



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Reply via email to