SGI: Hardware

First post, problem with one of the harddisks installed in an "Octane 2"

A few months ago I have been given an Octane 2, still in perfect condition. One thing that has been bugging me is, first thing when I turn it on, it gets past the diagnostics (and beeps, don't know if this is good or not), then it reports that some filesystems on /disk2 failed to mount. It turned out it has to do with /disk2 not being found. Computer then boots just fine, and I can still access disk2 and create directories. Weird thing is, after some further investigation I've found that the secondary hard disk isn't even powering on. I'm new to UNIX in general, so I have no idea what's happening. Is the hard disk faulty? I've tried putting it into another slot and that unfortunately hasn't fixed the problem.

I think I might be getting another Octane soon, and I will be able to further test the hard disk(s).
ManwelB wrote: I can still access disk2 and create directories. Weird thing is, after some further investigation I've found that the secondary hard disk isn't even powering on.


Looks like the HDD is dead while you access just its mount point at /disk2
What does the content of the file /etc/fstab say?
Project:
Temporarily lost at sea...
Plan:
World domination! Or something...
ManwelB wrote: I can still access disk2 and create directories

you can't, unix mounts are "seemless". that means you just create a folder somewhere and mount your device "into" that.
if the device is missing the folder is still there and you can use it but it's stored on your root disk instead.
r-a-c.de
foetz wrote:
ManwelB wrote: I can still access disk2 and create directories

you can't, unix mounts are "seemless". that means you just create a folder somewhere and mount your device "into" that.
if the device is missing the folder is still there and you can use it but it's stored on your root disk instead.

Oh ok, I see now.

vishnu wrote: What does the content of the file /etc/fstab say?

I can see this in my fstab file:
/dev/root / xfs rw,raw=/dev/rroot 0 0
/dev/dsk/dks0d250 /disk2 xfs rw 0 0

When executing the "hinv" command, this is what I get:
Image
It's not detecting the hard disk, obviously.

Call me stupid, but I think a faulty hard disk would usually power on just fine, the only thing is you wouldn't be able to access it. I've checked its pins and exterior to check for any damage or discoloration on the PCB, seems to be OK. I'd love to have a secondary hard disk, the first one is full up with only 2GB of available space, and I would really love to make an FTP server, but it seems like I'm not going to be doing it :?
What's the physical layout of the disks in the case? There are three slots for drives - which is being used by the system disk and which by disk2? It may be that disk2 has been moved from middle to top and the partition being seen by the system as /dev/dsk/dks0d3s0, though it should still show up in hinv as "unit 3" on controller 0. I'm assuming that your PROM boot variable is still pointing to disk 1, the lowest slot.

Same if someone has repartitioned disk2 as an option disk - /dev/dsk/dks0d2s0 will be pointing to a partition that no longer exists (but again, if disk2 were there hinv would see it whether it has a partition 0 or partition 7).

Note - you post state fstab to contain "/dev/dsk/dks0d250" - is that a copy-and-paste or a tr-yping typo? It ought to read "/dev/dsk/dks0d2s0" (s0 = partition 0 on d2 = disk with SCSI ID2 on dks0 = controller 0).

Can you hear two disks spinning up / "chattering"? (maybe not over Octane fan sound!) or, if you take the good disk out, can you here the other one? (IRIX won't boot, of course, if you take the good system disk out).
Fuel ; Indigo2 ; Octane ; RiscPC Kinetic/448MB/RISCOS4.39 or Debian-etch; EspressoPC ViaC3/900MHz/256MB/Debian-testing; RPi B RISCOS5.23; Rpi2 Raspbian-jessie; A5000/33MHz/FPA11/8MB/RISCOS3.11; A540/25MHz/FPA10/16MB/RISCOS3.11 or RISCiX1.21; R140/35MHz/4MB/RISCOS3.11 or RISCiX1.21
ajw99uk wrote: What's the physical layout of the disks in the case? There are three slots for drives - which is being used by the system disk and which by disk2? It may be that disk2 has been moved from middle to top and the partition being seen by the system as /dev/dsk/dks0d3s0, though it should still show up in hinv as "unit 3" on controller 0. I'm assuming that your PROM boot variable is still pointing to disk 1, the lowest slot.

Same if someone has repartitioned disk2 as an option disk - /dev/dsk/dks0d2s0 will be pointing to a partition that no longer exists (but again, if disk2 were there hinv would see it whether it has a partition 0 or partition 7).

Note - you post state fstab to contain "/dev/dsk/dks0d250" - is that a copy-and-paste or a tr-yping typo? It ought to read "/dev/dsk/dks0d2s0" (s0 = partition 0 on d2 = disk with SCSI ID2 on dks0 = controller 0).

Can you hear two disks spinning up / "chattering"? (maybe not over Octane fan sound!) or, if you take the good disk out, can you here the other one? (IRIX won't boot, of course, if you take the good system disk out).


Yes, it was indeed a typo.
The supposedly malfunctioning harddisk is installed in the middle slot. I tried installing it in the bottom slot, where the OS harddisk originally was but still, it did not power up. Not only can't I hear it spinning and working, but also unlike the OS harddisk, it's not heating up.
Sounds a bit buggered to me.
-----------------------------------------------------------------------
Hey Ho! Pip & Dandy!
:Octane2: :Octane2: :O2: :Indy: loft => :Indigo: :540: :Octane: :Octane: :Indy:
Indeed; might just be stiction if the machine has been sitting unused for a while, but I wouldn't want to rely on the disk (assuming it can be persuaded to spin again) after retrieving the contents.
Fuel ; Indigo2 ; Octane ; RiscPC Kinetic/448MB/RISCOS4.39 or Debian-etch; EspressoPC ViaC3/900MHz/256MB/Debian-testing; RPi B RISCOS5.23; Rpi2 Raspbian-jessie; A5000/33MHz/FPA11/8MB/RISCOS3.11; A540/25MHz/FPA10/16MB/RISCOS3.11 or RISCiX1.21; R140/35MHz/4MB/RISCOS3.11 or RISCiX1.21
ajw99uk wrote: Indeed; might just be stiction if the machine has been sitting unused for a while, but I wouldn't want to rely on the disk (assuming it can be persuaded to spin again) after retrieving the contents.


I have no means of accessing the hard disk outside of Octane. What do I do? I've swapped it several times, this is strange... I think it's destined to the dump.
Technically put, unless you give it a life saving jolt, aka dropping it off a table it really is FSCKED!

Years back I did have a drive I saved and in fact it lasted many years post. The heads did not like the edges of the platter, so I glued an eraser inside to stop it going to far, actually and I know I am old enough to suffer data loss in my memory, I'm sure the heads had actually fell off the platter, but that does seem a bit incredible. Anyhow, it worked what ever the issue was.
-----------------------------------------------------------------------
Hey Ho! Pip & Dandy!
:Octane2: :Octane2: :O2: :Indy: loft => :Indigo: :540: :Octane: :Octane: :Indy:
uunix wrote: Technically put, unless you give it a life saving jolt, aka dropping it off a table it really is FSCKED!

Years back I did have a drive I saved and in fact it lasted many years post. The heads did not like the edges of the platter, so I glued an eraser inside to stop it going to far, actually and I know I am old enough to suffer data loss in my memory, I'm sure the heads had actually fell off the platter, but that does seem a bit incredible. Anyhow, it worked what ever the issue was.


Ok, now I really think I messed up, for some reason, when I tried to shut down the Octane, it didn't shut down, I waited and waited, until my only option was to shut it down using the power button. When I turn it on, it now displays: PANIC: init died (why = 1, what = 0xd)
What the hell just happened?