qlogic driver , 1Tbyte hard error
scsi2 : QLogic QLA2200 PCI to Fibre Channel Host Adapter: bus 3 device 6 irq 5
Firmware version: 2.01.20, Driver version 4.15 Beta
Vendor: Diskbank Model: SANbank TLR-1000 Rev: 1.0
Type: Direct-Access ANSI SCSI revision: 02
scsi(2:0:0:0): Enabled tagged queuing, queue depth 16.
Attached scsi disk sdb at scsi2, channel 0, id 0, lun 0
SCSI device sdb: -1892903936 512-byte hdwr sectors (130345 MB)
sdb: unknown partition table
Signature · 2 lines
즐거운 하루 되세요. [email protected]
The maximum supported file system size under Linux 2.4 is just under 1Tb. The scsi layer gets slightly confused a bit earlier with its printk messages
Quoted 2 lines
The maximum supported file system size under Linux 2.4 is just under 1Tb. The scsi layer gets slightly confused a bit earlier with its printk messages
Is there any particular (that is, technical) reason for this? A few months ago I hit this problem while building a RAID system for our group. We wanted to do software RAID-0 over three hardware RAID-5 arrays (2 by 375G and one 525G) and the kernel (2.4.6) had a hissy fit.
Given the relatively low cost of disk space ($5000/terabyte and on up, see
http://staff.sdsc.edu/its/terafile/), is this something that will be supported in the future?
If you point me in the right direction I'd be willing to look at this issue.
Thanks
-Eric
Signature · 5 lines
-------------------------------------------- Eric H. Weigle CCS-1, RADIANT team [email protected] Los Alamos National Lab (505) 665-4937 http://home.lanl.gov/ehw/ --------------------------------------------
Quoted 3 lines
Given the relatively low cost of disk space ($5000/terabyte and on up, see
http://staff.sdsc.edu/its/terafile/), is this something that will be supported in the future?
2.5 Im sure
Quoted 1 line
If you point me in the right direction I'd be willing to look at this issue.
signed int overflow of 512 byte blocks
Quoted 8 lines · Eric Weigle
Quoted 2 lines
The maximum supported file system size under Linux 2.4 is just under 1Tb. The scsi layer gets slightly confused a bit earlier with its printk messages
Is there any particular (that is, technical) reason for this? A few months ago I hit this problem while building a RAID system for our group. We wanted to do software RAID-0 over three hardware RAID-5 arrays (2 by 375G and one 525G) and the kernel (2.4.6) had a hissy fit.
Quoted 5 lines
Given the relatively low cost of disk space ($5000/terabyte and on up, see
http://staff.sdsc.edu/its/terafile/), is this something that will be supported in the future?If you point me in the right direction I'd be willing to look at this issue.
There is a patch to allow 64-bit block devices - Ben LaHaise or Jens Axbow put it out about 2 months ago. I believe it fixes the SCSI midlayer, and a limited number of drivers. This is probably a good starting point (you may need to update the particular SCSI driver you are using).
Cheers, Andreas
--
Andreas Dilger \ "If a man ate a pound of pasta and a pound of antipasto,
\ would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/ -- Dogbert