No MB jumpers anymore thank god.
ata and scsi have drive jumpers for drive selection master/slave or 0-7 scsi..
sata drives also, but you position them via plugging into the MB controller 1-4 or 0-3 connectors typically as you know.
Their not all on the same cable as ata and scsi.
So sata provides an improved redundancy unlike scsi, all drives on one cable, where a bad cable can
wipe out an entire raid array.
I never did think much of the one cable concept with respects to raid redundancy.
The very thing raid is suppose to overcome (drive failure) and the cable becomes the weak link in the concept.
All bios have drive boot settings, so you could boot to physical drive 2 instead of drive 0.
Something to pay attention to, else you could wipe a good drive.
Thus the advent of the Rule of thumb and there's two way's to learn that rule...the easy way or the hard way...!!
The S/N is a definitive ID with respect to system reporting.
JFYI
1. Used "sfdisk ... > out; sfdisk .../sdc < out" to create matching partions on sdc, sda.
2. did NOT do the "dd..." to wipe the MBR, DID reboot after sfdisk just to be sure (and made sure fdisk -l showed valid partions on all drives - currently sda,b,c).
3. added sdc1 and sdc2 to md1 and md2 respectively, and letting the synch run.
Even though you did all that, raid still thought there was no valid boot sector and proceeded to sync the drives to
be identical to each other.
The raid system doesn't do a delete and re-init the partitions and copies files, it does a sector read/write/verify synced between the drives thru the controller.
It's faster and less CPU intensive.
fdisk will not remove/zero the boot sector when creating/deleting partitions, where as.... dd will.
The procedure calls for a blank drive because it's better to start with a zero'd out drive to reduce possible r/w/v errors
during the sync.
If Raid detects even one bit out of sync between drives it will force/flag a re-sync.
So it's not unusual for a sync to fail or a re-sync after a power outage even with a UPS.
The drives have to be bit for bit identical or it flags a sync.
In your case, the fact that raid began the sync says it didn't find a valid boot sector, else it would
not have started the sync.
Raid pretends to be smarter then an IT tech installing a drive with good data in a raid stack and won't harm it in any such event.
Tip
Keep in mind it only takes appox <10 millisec. for the system to render a good data drive unusable.
That's less time then it take's for one to say....Oh Sh_t...!!
I use the drink & think procedure, go get a drink and think about what your about to undertake.
Once you identify via S/N, then use a marker to ID the drive sda, sdb etc. good or bad.
Failure to ID can lead to further problems down the road.
You might have noticed most all drives have a barcode & S/N on them.
Good to make note of the number before install.
Else.... can you say Oh Sh_t...!!
Just so you know..... I've never have had to say.... Oh Sh_t...!
HaHa....
22 Replies
3814 Views