Re: Workarounds for advanced RAID features in GRUB Legacy?

[Leif W]

> From: "Tim Woodall" <address@hidden>
> Sent: 2005 September 11 Sunday 06:00


> I can't really see the point in mixing raid1 with anything else on a
> pair of disks but even so, I'd have

Oh yeah, certainly.  Sorry for the confusion.  I should elaborate. 
Though it's veering off the GRUB topic, this background info may apply. 
4 discs; 2xPATA @ 300GB, 2xSATA @ 400GB.  The RAID 0 partitions are the 
300GB partitions, the RAID 1 is the two RAID 0 partitions.  That leaves 
2xSATA discs with 100GB partitions, where I'll stuff as many OSes as I 
can and separate data spaces, RAID 1 when I can. :-)

At the least, I'm looking for two non-raided winxp installs (one on each 
disc).  If winxp can't play ball with the grown ups then it can go 
without redundancy.  ;-)  Besides it has a higher probability of 
breaking than my hardware, so I'd just be mirroring the same broken OS. 
They often require a reformat and reinstall after 6-12 months anyways. 
Then I'd like two installs of Linux, both RAID1, just incase I really 
mess one up and need something ready to go in a hurry.  RAID 1 will 
mirror the user error.  A separate, unmounted partition will not.  Two 
linux and two windows swap partitions (one on each disc).  And two data 
partitions per disc, Linux mirrored (2/3 remaining space), and WinXP 
just with two separate partitions, so that data (like software projects) 
doesn't live and die with the distribution.

Now, there may be things I overlooked, it's a work in progress, a hobby 
I've been nudging along for some time now.

> I don't run WinXP but this looks like a limitation in WinXP. If WinXP
> requires you to raid the entire disk then that is what you are going 
> to
> have to do for WinXP.

Well, I'm just basing my experience on the stock utilities in WinXP, and 
the disc management console didn't give me an option to select 
partitions to be part of a RAID array.  With the drivers and utilities 
for the SiI3112 chipset, WinXP can do only RAID 0 or RAID 1, only on 
SATA, and only the entire disc.  And it will happily attempt to make a 
RAID1 array even though it is running from the drive!  It's the computer 
equivalent of watching something so mind bogglingly stupid like Jack.*s 
that you can only laugh.  Actually I might end up keeping that OS on 
another system with very similar hardware.

> I suspect, however, that if you raid the whole
> disk but then in linux treat the end of each disk as non-raid and then
> run software raid in linux it will still all work correctly. (for 
> raid1
> at least although you definitely want to check that rebuilding the
> mirror in windows doesn't break the linux raid)

If I had more time and energy I'd be curious to see the results myself. 
Probably worth a good laugh at worst if it fails miserably, and at best 
utter astonishment if nothing broke.

> It's not unusual when a disk fails, expecially IDE disks, for it to
> crash the machine hard so you are going to have to reboot. And it's 
> also
> not unusual for the bios to crash if there is a failed disk meaning 
> you
> can't boot until you unplug the failed disk. (Almost all my 
> experiences
> of failed disks have either been bad sectors (which hopefully raid1 
> will
> compensate for) or failed to start after a shutdown (where bios 
> hanging
> isn't uncommon) - but it's not like I'm in a datacentre (which would 
> all

My experience was most likely due to thermal issues.  Despite being 
concerned at the volume (and rotational torque ;) of my case due to the 
15 or so fans and specialized heat sinks, several discs died during some 
hot weather with no A/C, and I didn't chalk it up to coincidence.  The 
system didn't crash hard.  The discs would start knocking audibly, and 
access times crawled.  After a while it was recognized by the BIOS but 
not the OS, and then no longer recognized by the BIOS, like nothing was 
there.  Air flow and temperature monitoring are more important to me 
now, so is some redundancy and external backups, so I can at least limp 
along.

> If you are very lucky the machine will stay up, mdadm will flag the 
> disk
> bad and you will get a message. A bit less lucky and the machine will
> crash but a power cycle will be sufficient. Worst case, you will have 
> to
> unplug the cable from the failed disk in order to get everything up
> again. Once you have the new disk you can install inplace of the 
> failed
> disk and rebuild the raid (or have another spare)

Well if that's the way it is, I can understand.  So long as it's not 
something that can be configured away.

Thanks again for the commentary.


Leif