Hi, On Sun, Aug 21, 2005 at 02:24:57PM -0700, darin strait wrote: > > The system was equipped with its original IDE disk and a newer 200GB > > one (btw: all 200gigs are available with newer 2.6 kernels). > I had gotten an email off-list from someone else who had put a large > disk into his system and had it work. I had thought that the drive size > limitation was a hardware problem (a lack of addressing bits), at least > with x86 PC hardware. Am I wrong, or does it not matter somehow since > this is Sun hardware? I have a 160GB IDE disk working flawlessly on an Ultra10 (well... I begin not to stand anymore the noise of the fans of this host). :-) hdb: Maxtor 6Y160P0, ATA DISK drive hdb: 320173056 sectors (163929 MB) w/7936KiB Cache, CHS=19929/255/63, (U)DMA > (fwiw, I would like to stick with 2.4 primarily because 2.6 has a bad > bug involving the parallel printer port. I believe that this bug is > still open.) # uname -r 2.4.27-2-sparc64 > The old 9 GB IDE was giving me trouble last week (read errors), so I > swapped a spare 120 GB I had from another project into the U10. I was > getting around 12.5 mb/s with the old disk, now I get about 15 mb/s > with the 120 GB disk (which is a bog-standard,n not-a-speed-demon > disk). It's noticably quicker, but I'm still interesting in size rather > than speed. I won't be able read or write files faster than 100 mbit > ethernet speeds, regardless of what sort of disk(s) are in the server. # hdparm -tT /dev/hdb /dev/hdb: Timing O_DIRECT cached reads: 640 MB in 2.00 seconds = 320.00 MB/sec Timing O_DIRECT disk reads: 46 MB in 3.02 seconds = 15.23 MB/sec Best regards, Frederic Lehobey
Attachment:
signature.asc
Description: Digital signature