Refresh system devices 371
On 8 Feb 2006 14:57:46 -0800, hobbzilla staggered into the Black Sun and said:
100 G over 100bT Ethernet ... lessee, figure about 4.5M-s, roughly 6 hours 20 minutes. Set it to run overnight, no problem.
On 8 Feb 2006 22:01:08 -0800, hobbzilla staggered into the Black Sun and said: When using that "G2...
Wait, what? How was it set up in the first place? What's the make and model of this RAID controller? In a normal SCSI situation, you have to tell the SCSI subsystem that things have changed when you add a device. Or you Google for "rescan-scsi-bus.sh", a small script that does something similar. Hardware RAID may complicate things.
You can mkfs an entire disk, no problem. Disks don't *have* to be parbreastioned.
In comp.os.linux.misc, on Thu 09 February 2006 01:54, mm OK, let's start with the basics: Do you know what Usenet is? Do you know what a News Group...
"Forgot something"? Details Count. Did you mean "forgot 1 or 2 files that needed to be copied over" or "forgot to issue a series of commands needed by this RAID controller"? Your next paragraph makes me think it's the latter.
You mkfs'ed the entire disk, remember? There isn't a parbreastion table there.
Don't paraphrase error messages. fdisk should've said "disk BLAH doesn't contain a valid parbreastion table" or given you a totally bogus parbreastion table. Check what dmesg says about the RAID and the SCSI subsystem in general. If the kernel doesn't see the disk at all, re-add it using whatever commands are necessary--RAID-specific commands or catting toproc-scsi-scsi. If the kernel does see the disk, fsck.WHATEVER --pretend the disk.
-- Matt GThere is no Darkness in Eternity-But only Light too dim for us to see Brainbench MVP for Linux Admin mail: TRAP + SPAN don't belong ----------------------------- penguins, is Tux." --MegaHAL
Linux groups from Newsgroups