headers
Kevin Lahey | 5 Aug 2002 18:45

Cobalt Qube 2 power supply problems?


Well, my poor hot-rodded Cobalt Qube 2 finally blew a gasket.
I shouldn't be too surprised -- I'd upgraded from 32MB of RAM
to 128MB, from 6GB of disk to 64GB, and added a wireless PCI 
card to make it into a host access point.  All that *had*
to have some impact.

The problems started a few days ago, when I woke up to find
the box unresponsive.  I checked the serial console, and it
continuously complained about errors writing to disk.
When I turned it off and then on again, it failed to read
the kernel from the boot disk.

I had visions of losing my (un-backed-up!) mail and digital
pictures, but then I found that the disk worked on my i386 
box.  Argh!  And the Qube would boot off of the original
6GB disk.

I tried some more experiments, to see if the boot image on
the 64GB disk was bad.  I found that the Qube would power
down as I tried to copy new files onto the original 6GB disk.
It just shut itself off!

I removed the PCI wireless card, and it stayed up a little
longer.  While toggling power I saw the fan surge on and off
in about three second cycles, in the way that I saw it surge
once before, when the power adaptor went out.

Considering all the evidence, it sounds like the power is 
bad.  Do y'all have any suggestions?  I don't mind springing
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bob Bostwick | 6 Aug 2002 23:34

?fdisk?

Hello,

	I've been trying to install NetBSD 1.5.3 on my RAQ2 for about 3
weeks now.  I have an OpenBSD box setup running NFS server, and can get
the RAQ2 to net boot to it. I then try to setup the drive for NetBSD,
but every time I try *desperate laugh*, I get a drive looking like
this...

NetBSD disklabel disk geometry:
cylinders: 16383 heads: 16 sectors/track: 63 (1008 sectors/cylinder)

BIOS disk geometry:
cylinders: 16383 heads: 16 sectors/track: 63 (1008 sectors/cylinder)

Partition table:
0: sysid 131 (Linux native)
    start 1, size 1024127 (500 MB), flag 0x0
        beg: cylinder    0, head   0, sector  2
        end: cylinder 1015, head  15, sector 63
1: sysid 130 (Linux swap or Prime or Solaris)
    start 1024128, size 263088 (128 MB), flag 0x0
        beg: cylinder 1016, head   0, sector  1
        end: cylinder 1023, head  15, sector 63
2: sysid 169 (NetBSD)
    start 1287216, size 410256 (200 MB), flag 0x0
        beg: cylinder 1023, head  15, sector 63
        end: cylinder 1023, head  15, sector 63
3: sysid 169 (NetBSD)
    start 1697472, size 23732352 (11588 MB), flag 0x0
        beg: cylinder 1023, head  15, sector 63
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kevin Lahey | 7 Aug 2002 01:26

Re: ?fdisk?

In message <003001c23d91$0469f3c0$440a010a <at> lnk.digitechsystems.com>,
"Bob Bostwick" writes:
>
>	I've been trying to install NetBSD 1.5.3 on my RAQ2 for about 3
>weeks now.  I have an OpenBSD box setup running NFS server, and can get
>the RAQ2 to net boot to it. I then try to setup the drive for NetBSD,
>but every time I try *desperate laugh*, I get a drive looking like
>this...
>
>Partition table:
>0: sysid 131 (Linux native)
>    start 1, size 1024127 (500 MB), flag 0x0
>        beg: cylinder    0, head   0, sector  2
>        end: cylinder 1015, head  15, sector 63
>1: sysid 130 (Linux swap or Prime or Solaris)
>    start 1024128, size 263088 (128 MB), flag 0x0
>        beg: cylinder 1016, head   0, sector  1
>        end: cylinder 1023, head  15, sector 63
>2: sysid 169 (NetBSD)
>    start 1287216, size 410256 (200 MB), flag 0x0
>        beg: cylinder 1023, head  15, sector 63
>        end: cylinder 1023, head  15, sector 63

Here's your problem.  You can have only one NetBSD FDISK slice
on the disk.  You should then carve up that FDISK slice into
the different filesystems/partitions for NetBSD, as you have
successfully.  If you set the sysid to something besides 169
or 165 (the old sysid for NetBSD) in slice 2, you should be in 
better shape.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Miller, Eoin | 7 Aug 2002 19:32

newbie trying to get NetBSD 1.5.3 on Qube 2700

hello all, i am attempting to get NetBSD 1.5.3 to run on my Cobalt Qube
2700, i have setup and exported the root directory for the qube via NFS and
setup dhcp without issue, the qube boots from the network, then it starts
reading from the nfs share and it looks for /boot/vmlinux.gz, but from what
i have read it is supposed to look for a different gzip'd kernel,
vmlinux_raq-2800.gz or something the like, i have followed the instructions
on this write up:

http://mail-index.netbsd.org/port-cobalt/2001/10/01/0001.html

and i have also tried the diskless FAQ and the FAQ by NetBSD for setting up
NetBSD on the Qube but i have had no luck, i tried nenaming the kernel to
/boot/vmlinux.gz and it does indeed seem to copy the file from the NFS
share, i checked this by using tcpdump, and after i renamed the file instead
of only 7 packets being passed between hosts 1597 packets are passed between
hosts, is there a step i am missing? please let me know.

thanks

**********************************************************************
This email and any files transmitted with it are confidential and are 
intended solely for the use of the individual or entity to whom it is 
addressed.  This communication may contain protected or privileged 
material and should only be viewed by the intended recipient(s).  If 
you are not the intended recipient or the person responsible for 
delivering the email to the intended recipient(s), be advised that 
you have received this email in error and that any use, dissemination, 
forwarding, printing or copying of this email is strictly prohibited.
**********************************************************************
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bob Bostwick | 8 Aug 2002 22:40

/: bad dir ino 2 at offset 0: mangled entry

First thanks for all the help with fdisk, and disklabel.  I managed to
get it to boot, and all was looking great.  I was still in single user
mode at this point, so I net-booted once more, and set rc.conf up for
multiuser.  Once I re-booted I got the message below.  I have tried
fsck, but it didn't seem to help.  I am still a newbie, so I'm sure it's
something simple I missed.

Boot device: wd0
root on wd0a dumps on wd0b
root file system type: ext2fs
/: bad dir ino 2 at offset 0: mangled entry
panic: bad dir
Stopped in init at      0x801224d4:     jr      ra
                bdslot: 0x801224d8:     nop
db> trace
801224d0+4 (ff01,0,d,0) ra 800431c8 sz 0
8004311c+ac (801568d0,0,72,0) ra 80102e18 sz 48
80102dc4+54 (801568d0,0,8015cee8,0) ra 800e5bd0 sz 32
800e57ec+3e4 (801568d0,0,8015cee8,c913dea0) ra 80063ccc sz 152
80063888+444 (801568d0,0,8015cee8,c913dea0) ra 80063788 sz 96
800633a0+3e8 (801568d0,0,8015cee8,c913dea0) ra 8002611c sz 120
main+92c (801568d0,0,8015cee8,c913dea0) ra 13f sz 88
PC 0x13f: not in kernel space
0+13f (801568d0,0,8015cee8,c913dea0) ra 0 sz 0
User-level: pid 1
db>3, head  15, sector 63

Thanx

B
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kevin Lahey | 8 Aug 2002 23:14

Re: /: bad dir ino 2 at offset 0: mangled entry

In message <000601c23f1b$c999c540$440a010a <at> lnk.digitechsystems.com>,
"Bob Bostwick" writes:
>First thanks for all the help with fdisk, and disklabel.  I managed to
>get it to boot, and all was looking great.  I was still in single user
>mode at this point, so I net-booted once more, and set rc.conf up for
>multiuser.  Once I re-booted I got the message below.  I have tried
>fsck, but it didn't seem to help.  I am still a newbie, so I'm sure it's
>something simple I missed.

What does your partition table look like?  Could you have accidently
set up wd0a as the linux /boot partition?  When the system boots,
it expects wd0a to contain the NetBSD / partition.  As an alternative,
perhaps you got wd0a set up correctly, but the disklabel claims it
is ext2fs instead of ffs?

Good luck,

Kevin
kml <at> patheticgeek.net

[I'm still waiting on my $90 power supply replacement from Sun, so it
might be awhile before I'm running port-cobalt again...]
Permalink | Reply | 1 comment |
headers
Peter H. Froehlich | 8 Aug 2002 23:34
Picon

Power Supply...

Hi there!

On Thursday, August 8, 2002, at 02:14 , Kevin Lahey wrote:

> [I'm still waiting on my $90 power supply replacement from Sun, so it
> might be awhile before I'm running port-cobalt again...]

In January 2002, when I asked Sun for a power supply, they told me 
that they don't have any left. Robert Thille then pointed me here:

http://www.alliedelec.com/cart/partlookup.asp?ST=AS&PartNumber=688-5000

I ordered my power supply from these guys for about $50 and it 
works fine (well, last I checked it worked fine, haven't powered 
the Qube up in a while). Maybe some kind person should add this to 
the FAQ?

Peter
--
Peter H. Froehlich []->[!]<-[] http://nil.ics.uci.edu/~phf/
OpenPGP: D465 CBDD D9D2 0D77 C5AF  353E C86C 2AD9 A6E2 309E
Permalink | Reply | 0 comments |
headers
Kevin Lahey | 9 Aug 2002 00:44

Re: Power Supply...

In message <9C433E92-AB16-11D6-A2E0-0003936CFB76 <at> ics.uci.edu>"Peter H. Froehlic
h" writes
>Hi there!
>
>On Thursday, August 8, 2002, at 02:14 , Kevin Lahey wrote:
>
>> [I'm still waiting on my $90 power supply replacement from Sun, so it
>> might be awhile before I'm running port-cobalt again...]
>
>In January 2002, when I asked Sun for a power supply, they told me 
>that they don't have any left. Robert Thille then pointed me here:

The Sun Store claimed to have 'em in stock, but they also claimed it
might take 16 days to ship it.  Yeesh.  An alternative would
certainly be welcome!

>http://www.alliedelec.com/cart/partlookup.asp?ST=AS&PartNumber=688-5000
>
>I ordered my power supply from these guys for about $50 and it 
>works fine (well, last I checked it worked fine, haven't powered 
>the Qube up in a while). Maybe some kind person should add this to 
>the FAQ?

It looks like they might be out of stock on that.  The website
says something about not back-ordering it, either.  I'd certainly
like to have more little Cobalt details in the FAQ;  the Linux
FAQs are pretty decent, but don't seem to have all of these
important little bits of information.

Thanks!
(Continue reading)

Permalink | Reply | 0 comments |
headers
John Mackey | 9 Aug 2002 07:51
Picon
Favicon

Problems booting Cobalt Qube 2

Hi ppl,

I have been struggling to install NetBSD on my cobalt Qube 2 .

I have followed the steps on http://www.netbsd.org/Ports/cobalt/faq.html.

Some details on my configuration 

# fdisk
NetBSD disklabel disk geometry:
cylinders: 8912 heads: 15 sectors/track: 63 (945 sectors/cylinder)

BIOS disk geometry:
cylinders: 8912 heads: 15 sectors/track: 63 (945 sectors/cylinder)

Partition table:
0: sysid 131 (Linux native)
    start 63, size 31328 (15 MB), flag 0x0
        beg: cylinder    0, head   1, sector  1
        end: cylinder   33, head   3, sector 17
1: sysid 169 (NetBSD)
    start 42336, size 8379504 (4091 MB), flag 0x0
        beg: cylinder   44, head  12, sector  1
        end: cylinder  719, head  14, sector 63
2: <UNUSED>
3: <UNUSED>

partition> E
# /dev/rwd0d:
type: ESDI
(Continue reading)

Permalink | Reply | 3 comments |
headers
Ted Unangst | 9 Aug 2002 09:23
Picon
Favicon

Re: Problems booting Cobalt Qube 2

On Fri, 9 Aug 2002, John Mackey wrote:

> I have put this  kernel
>
> into the /boot directory and renamed in to vmlinux_raq-2800.gz

that should be /stand/boot in your case.  /boot on the linux paritition.
also make sure your filesystem is the old ext2fs, not version 2 or
whatever the default is.

--
"The laws in this city are clearly racist.  All laws are racist.
The law of gravity is racist."
       - M. Barry, Mayor of Washington, DC
Permalink | Reply | 1 comment |

Gmane