[fo0bar]

Greetings! First time poster, I've got an SGI Challenge S (Indy without the graphics hardware) and just received a SCSI2SD V6, and I'm trying to install IRIX 6.5.22 on it.

fx works successfully, and sees sane disk sizes.

Code:

----- partitions-----
part  type        blocks            Megabytes  (base+size)
  0: xfs      266240 + 29093888    130 + 14206
  1: raw        4096 + 262144        2 + 128
  8: volhdr        0 + 4096          0 + 2
10: volume        0 + 29360128      0 + 14336

capacity is 29360128 blocks

However, miniroot/unix.IP22 ends with:

Code:

>> bootp():i/22/ovl1/dist/miniroot/unix.IP22
Setting $netaddr to 10.9.8.31 (from server 10.9.8.54)
Obtaining i/22/ovl1/dist/miniroot/unix.IP22 from server 10.9.8.54
Setting $netaddr to 10.9.8.31 (from server 10.9.8.54)
Obtaining i/22/ovl1/dist/miniroot/unix.IP22 from server 10.9.8.54
IRIX Release 6.5 IP22 Version 10070055 System V
Copyright 1987-2003 Silicon Graphics, Inc.
All Rights Reserved.

WD95A SCSI controller 4 - differential external, rev 0, min xfer period 100ns
WD95A SCSI controller 5 - differential external, rev 0, min xfer period 100ns
gfe1: missing
gfe0: missing
xpi0: missing from slot 1
WARNING: Kernname environment variable not set by sash. Runtime symbol table not loaded. Loadable modules will not be registered or loaded.
WARNING: initial mount of root device /hw/node/io/gio/hpc/scsi_ctlr/0/target/1/lun/0/disk/partition/1/block failed with errno 1011
PANIC: vfs_mountroot: no root found 
[Press reset to restart the machine.]

(A full fx / miniroot log is attached, as well as SCSI2SD XML.)

SCSI2SD V6has latest firmware (v6.3.2) and is set up as so: No terminator, no limit, parity, SCSI2, short SCSI selection pulses, 6 second startup delay.  Hard drive, ID 1, start 0, sector size 512, sectors 29360128, default vendor etc. However, I've basically played with flipping all of these options without any meaningful change.

I can't find anything about "errno 1011".  The kernname warning doesn't make any sense to me as the miniroot hasn't even been installed yet, but I could be wrong (kernname indeed isn't set, but I'm not sure what I would set it to).  Env info:

Code:

>> printenv
AutoLoad=Yes
TimeZone=PST8PDT
console=d1
diskless=0
dbaud=9600
volume=80
sgilogo=y
autopower=y
netaddr=10.9.8.31
eaddr=08:00:69:0b:63:59
ConsoleOut=serial(0)
ConsoleIn=serial(0)
cpufreq=180
SystemPartition=scsi(0)disk(1)rdisk(0)partition(8)
OSLoadPartition=scsi(0)disk(1)rdisk(0)partition(0)
OSLoadFilename=/unix
OSLoader=sash
>> hinv
                  System: IP22
                Processor: 180 Mhz R5000, with FPU
    Primary I-cache size: 32 Kbytes
    Primary D-cache size: 32 Kbytes
    Secondary cache size: 512 Kbytes
              Memory size: 256 Mbytes
                SCSI Disk: scsi(0)disk(1)

Hardware is known good; I had a Debian installation on a 9GB SCSI drive, before swapping out for the SCSI2SD.

Any ideas? Thanks.

[robespierre]

Here is the main issue:

Code:

PANIC: vfs_mountroot: no root found

Warnings are things to take note of, but panics are fatal errors that force the machine to halt.

The miniroot must be copied to a partition of a local disk*. Not by IRIX, but by the standalone environment.

This copy can be carried out through the PROM GUI, by selecting "2. Install System Software" from the boot menu. It can also be accomplished from the PROM command line.

* Unless the client and server are configured for "diskless" operation. See https://techpubs.jurassic.nl/manuals/053.../ch01.html for details.
You have the diskless PROM variable set to 0, so without a local root volume it expects, IRIX panics.

[fo0bar]

Thanks for the info! I was going off a guide which implied both the PROM's "Install System Software" option and booting miniroot/unix.IP22 from overlay disc 1 were equivalent, and either would bootstrap the miniroot.

Complicating matter is a bit odd, it appears the I/O is unstable (SCSI timeouts) if the SCSI2SD does NOT have termination enabled, whereas everything I've read seems to say you definitely don't want termination on the drive because the internal ribbon has a terminator (which it does on my system). I checked with the original hard drive and it is jumpered for its own termination, so at least the SCSI2SD config is now consistent with the drive. Still, huh!

Currently in progress installing IRIX 6.5.22, thanks!

[robespierre]

Complicating matter is a bit odd, it appears the I/O is unstable (SCSI timeouts) if the SCSI2SD does NOT have termination enabled, whereas everything I've read seems to say you definitely don't want termination on the drive because the internal ribbon has a terminator (which it does on my system). I checked with the original hard drive and it is jumpered for its own termination, so at least the SCSI2SD config is now consistent with the drive.

The terminator on your ribbon cable may have failed! It could also be a lousy passive type. Was it original with this system?
Also, terminators that are external (not built in to a device) rely on the TERMPWR line. Since the SCSI2SD is also connected to TERMPWR it may be loading it down or failing to supply it. I'm not familiar enough to say. Your original disk could have had the same problem.

Currently in progress installing IRIX 6.5.22, thanks!

Cool, y/w!

[fo0bar]

The terminator on your ribbon cable may have failed! It could also be a lousy passive type. Was it original with this system?
Also, terminators that are external (not built in to a device) rely on the TERMPWR line. Since the SCSI2SD is also connected to TERMPWR it may be loading it down or failing to supply it. I'm not familiar enough to say.

Here's what it looks like -- install is still in progress and the SCSI2SD is precariously balancing during it, so that's as far as I'm handling it until install is done. Oh well, now that it works, I'm content that it does.

I originally got this Challenge S second-hand in 2002; the setup was the same now as then, but with the 9GB drive instead of the SCSI2SD. No idea if that's how the previous owner got it from SGI.