Boot problem

Hi,

I have an Archimedes 410/I RISC OS3 which has been in storage for almost 2 years. It fires up, red screen, blue screen, SCSI located and I get the basic icon bar but the rest of the screen stays blank. ie: it is not Booting up my last left Desktop. Clicking on the SCSI drive fails to Boot up the Desktop. F12 *St. shows the correct configuration.
Can anyone guide me into getting the Desktop to Boot-up please ?
Very Many Thanks,

Gary

It fires up, red screen, blue screen, SCSI located and I get the basic icon bar but the rest of the screen stays blank. ie: it is not Booting up my last left Desktop. Clicking on the SCSI drive fails to Boot up the Desktop.

Are you able to access all the file on the SCSI drive? If so are you sure the status readout shows Boot (as opposed to NoBoot) that the boot filesystem is SCSI and that the boot options are set correctly.
In the course of 2 years the CMOS settings will have been lost.

Hi Steve,
Thanks for getting back to me.
If I click on the SCSI drive icon I get the hour-glass running before the message ‘Unable to mount this medium’ so, No, I cannot at this time access the SCSI drive files.
The ST. configuration definitely shows Boot (not NoBoot).
???
Gary

If I click on the SCSI drive icon I get the hour-glass running before the message ‘Unable to mount this medium’ so, No, I cannot at this time access the SCSI drive files.

Normally SCSI interfaces are the bit that responds to the system in a way that you see at startup¹, but the drive itself may actually have failed. Your missing boot sequence is on the SCSI and all you are seeing is the ROM based boot items.

I think you need some of the RISC OS SCSI diagnostics stuff on floppy that came with the SCSI drive.

¹ Across the board, PC hardware too.

Thanks Steve,

That sounds about right. Not sure what disks I have back in the attic but will get up there this week-end and see what I can find. Will let you know the outcome.
Thanks for your very much appreciated help!

Best,
Gary

Don’t need discs for the initial tests…

Can you hear your drive spin up?

Get to where your drive is, pull your hair out of the way, and try to get your ear as close to the drive as possible.

This is what it ought to sound like:
fweeee bzzzt chk chk chk bzzzt

This is bad:
fweeee bzzzt chk bzzzt chk bzzzt chk [repeating]

Absolute silence is also bad.

Explanation:
fweeee – drive spinning up
bzzzt – head moving a long distance across the disc
chk – head stepping to a nearby track

In the first example, the good one, the drive spins up, moves from its parking position to read a few sectors of data (its own self-configuration), and then takes the head to wherever it rests when waiting for something to do.

In the second, bad example, the drive spins up and takes the head from parking position to try to read its housekeeping data, but for some reason it cannot so it will try to home the head and step it out again, and again, and again…
Some drives may give up and spin down after so many tries. Some will keep on trying, forever.

An old SCSI drive is probably too old to do a thermal calibration. How this sounds depends upon the drive, it can range from a soft “chk-chk-chk-boink” to a louder “shee-shee-clunk” (I swear it was slamming the head into the end stop to knock dirt off or something).

Assuming your drive is powered up and is rotating and isn’t sounding broken…

Time to see how much the computer can ‘see’.

Start with *SCSI:Devices and see if anything shows up.
If it does, then do a *SCSITest (if available) to verify that the SCSI card is working correctly.

Then… You don’t say what sort of SCSI interface you have. If it is a Morley one, you need to use *Configure SCSIFSLink <drive> <scsi_id> to associate SCSI drive <scsi_id> with FileCore drive <drive>.

After that, your drive should be visible to the system.

If the drive appears to be spinning but *Devices did not list it, it is time to drag out the crystal balls. You’ll need to check the wiring, the terminators, and so on. Ideally if the machine has not been touched at all, it may just be a matter of unplugging everything and plugging it back in. But if that doesn’t work…… all I can say is “good luck!”.

Hello Rick,

Thanks for diving in with me.
Ok,seems the sound effects are right and the (Morley) SCSI drive spins up nicely. Devices shows
Device Type Capacity Vendor Product Rev Trans
0 Direct Access Unknown Fujitsu Not ready 0485 Async
7 Host(uncached) Morley SCSI Expansion 1.19

SCSITest shows
Card:0 DMA Counter(runs up the numbers) :Pass FIFO :Pass Cache :None

The SCSIFSLink shows 0:— 1:— 2:— 3:— 4:0 5:— 6:— 7:—
This spec has always worked fin in the past so I am a little reluctant to change it – unless you feel certain I should. The SCSI drive number is 4.

As the system can clearly ‘find’ the SCSI drive I am suspecting it is just something in the Boot file which is failing to link to the SCSI and show it.
Any further thoughts would be greatly appreciated ?
Meanwhile soon as ‘her in doors’ allows I will get back up into the attic to see what support discs I can find.
Thanks Again,

Gary

As the system can clearly ‘find’ the SCSI drive I am suspecting it is just something in the Boot file which is failing to link to the SCSI and show it.

Wouldn’t that be the “*Configure SCSIFSLink to associate SCSI drive with FileCore drive .” that Rick quoted?

curious: not ready?

Device Type Capacity Vendor Product Rev Trans
0 Direct Access Unknown Fujitsu Not ready 0485 Async
7 Host(uncached) Morley SCSI Expansion 1.19

I think the Fujitsu drive has shown ‘not ready’ since it was originally disconected when the SCSI drive went in. (Not too worried about this message.)
The good news is I now have the original Morley Electronics diskettes, Uitilities & Support discs 1 & 2 and also the Lingenuity SCSI Software Interface diskette plus another one simply marked SCSI. On this last diskette is a file called !BootSCSI but whn I click it I get the error message ‘Unknown or Missing Variable’.
SCSIFSLink is now set to 4 (same number as the SCSI drive) but still not finding it.
??

Also now getting a message Invalid SCSI device ID.
Any ideas how I find what my device ID is ?

Any ideas how I find what my device ID is ?

Normally set by links on the end of the drive.

Intermittent reading of SCSI information could be a loose SCSI cable terminator or a slight oxide build up on one of the connectors. As Rick said disconnect and reconnect the connectors (gives a small friction wipe of the metal-metal contacts)

Ok, this may be relevant.
The present configuration (F12 St.) is not showing any entry for SCSIFSHost.
Attempting *Configure SCSIFSHost 7 (or 4) (7 was in the original configuration)gives – ‘Bad Drive’
Guessing this may be the missing link – anyone got a *Configure command to re-instate ?

Thanks,
Gary

Done the cables and cleaned contacts etc. no change. (Thanx)

Ok, got the SCSIFSHost 7 showing. So configuration is now exactly as when system last running well.
Still seems to be booting-up in CMOS.

Sorry I’ve not been around. Tearing my hair out trying to work out why something useful like _kernel_RMAalloc() is in ANSILib (where I don’t need it) but not in Stubs (where it might, like, be useful!).

Aaaanyway: Looking at what you wrote, the SCSI host should be 7 (that’s normal, and correct). The assignment of device (#0) to drive (4) is correct.

As Steve indicated, the Not ready is saying that the drive is not currently ready to respond to requests… Therefore, it might have spun up and been ‘seen’ by the machine, but attempts to access it will fail because… it is… not ready. ;-)

[question is, though, why?]

I notice also that the drive capacity is “unknown”. This typically means that the drive’s SCSI interface has communicated with the host (so the computer knows something is connected) by the drive mechanism has not provided its data to pass on to the host.
It’s like the car has started but there’s nobody in the driver’s seat.¹

I’m not sure what to suggest next given that you say the drive is spinning up correctly…

If the drive had failed, it wouldn’t spin up correctly. If the power supply was weak, it shouldn’t spin up correctly.
Hmmm…

¹ One of my trademark naff analogies!

Therefore, it might have spun up and been ‘seen’ by the machine, but attempts to access it will fail because… it is… not ready. ;-)

Which is why I thought maybe there was a slightly iffy contact round the drive. A pullup/down on the ready sense line. Gary, that’s the links on the drive and the termination for the SCSI cable (normally at the opposite end of the cable to the host card) that need a lift and reseat.

EUREKA! Well.. almost.
Last night, having tried just about everything – with your splendid support, I had had enough, pushed the machine aside and went to bed. This morning I simply threw the switch (in ‘could no longer care’ mode)and – Bang! Worked first time – perfectly! Everything – exact.
After a few minutes however it gave me a low battery warning so I replaced the internal batteries and now, whilst all the boot-up sounds are there, the screen remains blank but with purple dots racing diagonally across the otherwise black screen. Weird, or what ?
Does anyone know how to correct this latest bizarre twist ?
Gary

I have checked the scart cable connections – seems good.
Re-starting brings up the correct boot noises but the screen still on purple racing flickers.
:(

After a few minutes however it gave me a low battery warning so I replaced the internal batteries

I’m confused I’m pretty certain that the only RISC OS computer that has a low battery warning is the Acorn A4!

If the 2 AA CMOS batteries loose their charge then you get anything from a blank screen to various error messages but none about the battery!
After changing a CMOS battery you need to do a ‘Delete’ Power ON to reset CMOS to sensible values.

Hi Chris,
I have an Archimedes 410/I which has 2 x AA batteries for the CMOS. Running a Morley SCSI and an Apps. pack that gives me helpful (low battery) messages. When replaced battteries before the Configuration is preserved. Problem I now face is with no readable screen even doing a CMOS reset won’t allow me to see anything so I will be unable to re-configure.
Can;t help but feel this has to be a vdu issue ???

Anyone know where I can buy a new Monitor/VDU for my Acorn ?
:)

tried keypad* + Reset and then type blind conf. sync 0

Thanks Wouter, have tried but the screen is high-speed scrambling black and purple flecks so no input is being seen. Nothing happened, have to consider the monitor has given up.
Quite happy to buy another one – if I can find it ?

Almost all monitors and most TVs with a Scart socket should work. The short thick Acorn 9pin DIN to Scart lead often ended up with broken wires.
We stock suitable cables and second Hand monitors:
http://www.cjemicros.co.uk/micros/prices/categories/monitors.shtml#shmonitor

I’m still confused about the battery message as there is no hardware connected capable of knowing if they are flat.

CMOS setting can be preserved if:
The settings are valid.
AND The batteries are changed whilst the computer is powered (This was against Acorns recommendation!)