I wrote:
> I'm therefore thinking of buying a removal disk for offsite
> backups.
> I am considering getting a Castlewood Systems (http://www.castlewood.com/)
> ORB 2.2 GB external SCSI drive. See a review at
> http://www8.zdnet.com/products/stories/reviews/0,4161,408937,00.html
>
> It's pretty cheap (US$230 and US$35 for extra disks) and seems
> very fast (11ms read/ write 12ms; 12.2 MB/sec max sustained
> transfer rate? Heck I don't get that from my Barracuda!)
I bought one from buy.com Tuesday evening and received it this
morning.
I got an external ultra-SCSI model (comes with 50 pin cable and a blank
2.2GB cartridge). It costs CDN$268 and I bought six blank cartridges for
CDN$39 each (1.45 $CND = 1 $US, so divide my prices by 1.45 to get US
dollars). Cartridges are 3 to 4 times cheaper than Jaz.
I plugged it in and booted. It detected okay (not sure about that reset):
Vendor: CWS ORB2 Model: -SE U ID 5 Rev: D20
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi removable disk sde at scsi0, channel 0, id 5, lun 0
(scsi0:0:5:0) Synchronous at 20.0 Mbyte/sec, offset 15.
scsi0 : channel 0 target 5 lun 1 request sense failed, performing reset.
fdisk is not too happy about the partition table:
Disk /dev/sde: 255 heads, 63 sectors, 268 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/sde1 1 261 2089024 5 Extended
Partition 1 has different physical/logical beginnings (non-Linux?):
phys=(1, 0, 1) logical=(0, 128, 1)
Partition 1 has different physical/logical endings:
phys=(517, 127, 63) logical=(260, 146, 14)
Partition 1 does not end on cylinder boundary:
phys=(517, 127, 63) should be (517, 254, 63)
/dev/sde5 1 261 2084480+ 83 Linux native
As you can see from the above, I decided not to delete the existing
partition, but simply tag it as ext2. I then created a filesystem on it
with:
# mke2fs -m 0 -i 16384 /dev/sde5
This gives me:
# df /dev/sde5
Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/sde5 2065354 2062069 3285 100% /orb
I copied data to it to fill it, and got an average of 2.1 MB/sec writing.
Reading is faster, of course:
# hdparm -tT /dev/sde5
/dev/sde5:
Timing buffer-cache reads: 64 MB in 1.95 seconds =32.82 MB/sec
Timing buffered disk reads: 32 MB in 3.91 seconds = 8.18 MB/sec
Am I happy with it? Yes.
Problems? Well, it could be because I now have 3 external SCSI devices and
too long of a chain, but I'm getting these messages logged:
scsi : aborting command due to timeout : pid 493526, scsi0, channel 0, id 0,
lun 0 Read (10) 00 00 76 97 ef 00 00 08 00
scsi : aborting command due to timeout : pid 493527, scsi0, channel 0, id 0,
lun 0 Read (10) 00 00 76 97 f7 00 00 10 00
SCSI host 0 abort (pid 493409) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
(scsi0:0:0:0) Synchronous at 20.0 Mbyte/sec, offset 15.
(scsi0:0:1:0) Synchronous at 20.0 Mbyte/sec, offset 15.
(scsi0:0:5:0) Synchronous at 20.0 Mbyte/sec, offset 15.
I'll report back if this turns out to be a serious problem related to the
ORB.
Peter
