[SunRescue] Bad - new NVRAM/TOD??

[steve price]

Hi all,

I recently pulled an old IPX and SS2 off the shelf and
fired them up. Of course, the ether address and
hostid
weren't there - so I replaced the NVRAM with a brand
new one from Mouser. Reset the hostid and ether
address.

Now at boot it complains that FAILURE TOD (f20007fa)
Bit failure, exp=000000ff,obs=0000007f. Can't op.
Power on Self Test Failed - Replace NVRAM.

However, the ether address and hostid remains after
removing power, and the clock keeps good time,
indicating the battery is ok.

I've pulled the sun-nvram-hostid.faq and reset - but
no avail.

Other than saying its failed and stopping the boot
process, the box is happy.

Any ideas on how to reset the bit it's looking for and
get a clean boot?

thanks!!



__________________________________________________
Do You Yahoo!?
Yahoo! Photos - Share your holiday photos online!
http://photos.yahoo.com/

[Harry Regan]

Hi Steve--

Yeah, I experienced the same thing and posted a message about it in December.
Since then, I've found out two additional data points. I originally ordered
200 NS chip for my IPX from Mouser and got the TOD error you reference. I
ordered a 70 NS version of the same chip (also from Mouser) and got the same
results.

Then I ordered the 150 NS version of the M48T12 from JDR Microdevices (they
don't carry the M48T02), and it worked like a champ. They're $16.99 each from
JDR, rather than $12.80 from Mouser for the M48T02. Mouser also carries the
M48T12 chip, also at $12.80, but I haven't tried it (yet).

So, despite what the NVRAM FAQ says, it looks like the MT48T12 chip is the one
to use. All the chips I've been playing with are from the same manufacturer,
so it's not a vendor problem.

The ones that did NOT Work:

Manufacturer: ST Label: M48T02-70PC1
Manufacturer: ST Label: M48T02-200PC1

The one that DID Work:

Manufacturer: ST Label: M48T12-150PC1

I will sware that I successfully used an M48T02 on an IPC a while back.
Anyone out there have an IPC with a dead NVRAM chip willing to experiment with
one of these chips? I'll send a free chip to the first one who volunteers!

P.S. I have not found a way around the TOD Clock error. I even tried some
kinky nvramrc gymnastics, but the machine would always blow back to the prom
prompt.



--- original message ---

Hi all,

I recently pulled an old IPX and SS2 off the shelf and
fired them up. Of course, the ether address and
hostid
weren't there - so I replaced the NVRAM with a brand
new one from Mouser. Reset the hostid and ether
address.

Now at boot it complains that FAILURE TOD (f20007fa)
Bit failure, exp=000000ff,obs=0000007f. Can't op.
Power on Self Test Failed - Replace NVRAM.

However, the ether address and hostid remains after
removing power, and the clock keeps good time,
indicating the battery is ok.

I've pulled the sun-nvram-hostid.faq and reset - but
no avail.

Other than saying its failed and stopping the boot
process, the box is happy.

Any ideas on how to reset the bit it's looking for and
get a clean boot?

thanks!!


___________________________________________________________________
---- Harry J. Regan
---- Phone: [phone]
---- FAX: [phone]
---- ***@usa.net
---- ***@lepus.com


____________________________________________________________________
Get free email and a permanent address at http://www.amexmail.com/?A=1

[r***@sunhelp.org]

> Hi Steve--
>
> Yeah, I experienced the same thing and posted a message about it in December.
> Since then, I've found out two additional data points. I originally ordered
> 200 NS chip for my IPX from Mouser and got the TOD error you reference. I
> ordered a 70 NS version of the same chip (also from Mouser) and got the same
> results.
>
> Then I ordered the 150 NS version of the M48T12 from JDR Microdevices (they
> don't carry the M48T02), and it worked like a champ. They're $16.99 each from
> JDR, rather than $12.80 from Mouser for the M48T02. Mouser also carries the
> M48T12 chip, also at $12.80, but I haven't tried it (yet).
>
> So, despite what the NVRAM FAQ says, it looks like the MT48T12 chip is the one
> to use. All the chips I've been playing with are from the same manufacturer,
> so it's not a vendor problem.
>
> The ones that did NOT Work:
>
> Manufacturer: ST Label: M48T02-70PC1
> Manufacturer: ST Label: M48T02-200PC1
>
> The one that DID Work:
>
> Manufacturer: ST Label: M48T12-150PC1
>
> I will sware that I successfully used an M48T02 on an IPC a while back.
> Anyone out there have an IPC with a dead NVRAM chip willing to experiment with
> one of these chips? I'll send a free chip to the first one who volunteers!

Harry,

I have several IPCs with dead NVRAM :^) .

I just installed one M48T02-200PC1 from Mouser. Haven't booted it yet. I
will report back to the list about the TOD results.

-- thanks, Steve Pacenka
email: ***@lightlink.com or ***@cornell.edu

[steve price]

harry, steve...

were the chips that worked (m48t12-150) used in an ipx
or ss2? they're the ones I have access to at the
moment, no ipc's.

I've actually found an old sun bug report (1055000)
that mentions the problem specifically on the 4/75
(ss2) as being a testing bit flipped by the
manufacturer, and not being reset properly by the
kick-start routine, and ... no resolution was
offered...

I wonder if we used an ipc to initially set the chip,
whether that would fix the issue?

Like you, I've tried several commercial sources and
have gotten the same chips from both - apparently
sgs/thompson is the buyer/successor to mostek.

I haven't tried any Dallas Semiconductor chips, mainly
since the nvram faq said they mostly don't work
either...

So I will also try the m48t12-150 - thanks for the
reply!!


__________________________________________________
Do You Yahoo!?
Yahoo! Photos - Share your holiday photos online!
http://photos.yahoo.com/

[Steve Pacenka]

> harry, steve...
>
> were the chips that worked (m48t12-150) used in an ipx
> or ss2? they're the ones I have access to at the
> moment, no ipc's.
>
> I've actually found an old sun bug report (1055000)
> that mentions the problem specifically on the 4/75
> (ss2) as being a testing bit flipped by the
> manufacturer, and not being reset properly by the
> kick-start routine, and ... no resolution was
> offered...
>
> I wonder if we used an ipc to initially set the chip,
> whether that would fix the issue?

Steve,

I'm imagining an IPC in continued loyal service, reviving NVRAMs for SS2s. :^)

If no one else advises against it, I will try this since I have an SS2 in the
rehab queue also, besides several IPCs. The first new NVRAM went into an IPC,
and after I have the IPC booting properly I will transplant the chip into the
SS2.

-- SP

[Mike Hebel]

RE: Test NVRAM chip...

Done! I'll volunteer for that!

Mike Hebel

-----Original Message-----
From: rescue-***@sunhelp.org [mailto:rescue-***@sunhelp.org]On
Behalf Of Harry Regan
Sent: Wednesday, January 03, 2001 7:17 AM
To: ***@sunhelp.org
Subject: Re: [SunRescue] Bad - new NVRAM/TOD??


Hi Steve--

Yeah, I experienced the same thing and posted a message about it in
December.
Since then, I've found out two additional data points. I originally ordered
200 NS chip for my IPX from Mouser and got the TOD error you reference. I
ordered a 70 NS version of the same chip (also from Mouser) and got the same
results.

Then I ordered the 150 NS version of the M48T12 from JDR Microdevices (they
don't carry the M48T02), and it worked like a champ. They're $16.99 each
from
JDR, rather than $12.80 from Mouser for the M48T02. Mouser also carries the
M48T12 chip, also at $12.80, but I haven't tried it (yet).

So, despite what the NVRAM FAQ says, it looks like the MT48T12 chip is the
one
to use. All the chips I've been playing with are from the same
manufacturer,
so it's not a vendor problem.

The ones that did NOT Work:

Manufacturer: ST Label: M48T02-70PC1
Manufacturer: ST Label: M48T02-200PC1

The one that DID Work:

Manufacturer: ST Label: M48T12-150PC1

I will sware that I successfully used an M48T02 on an IPC a while back.
Anyone out there have an IPC with a dead NVRAM chip willing to experiment
with
one of these chips? I'll send a free chip to the first one who volunteers!

P.S. I have not found a way around the TOD Clock error. I even tried some
kinky nvramrc gymnastics, but the machine would always blow back to the prom
prompt.



--- original message ---

Hi all,

I recently pulled an old IPX and SS2 off the shelf and
fired them up. Of course, the ether address and
hostid
weren't there - so I replaced the NVRAM with a brand
new one from Mouser. Reset the hostid and ether
address.

Now at boot it complains that FAILURE TOD (f20007fa)
Bit failure, exp=000000ff,obs=0000007f. Can't op.
Power on Self Test Failed - Replace NVRAM.

However, the ether address and hostid remains after
removing power, and the clock keeps good time,
indicating the battery is ok.

I've pulled the sun-nvram-hostid.faq and reset - but
no avail.

Other than saying its failed and stopping the boot
process, the box is happy.

Any ideas on how to reset the bit it's looking for and
get a clean boot?

thanks!!


___________________________________________________________________
---- Harry J. Regan
---- Phone: [phone]
---- FAX: [phone]
---- ***@usa.net
---- ***@lepus.com

[Harry Regan]

Steve--

The only machines I've tried this on are IPXes (IPXen? IPices?)...
But I have three to play with so I can demonstrate that the problem
in not unique to a particular machine...


--- original message ---

harry, steve...

were the chips that worked (m48t12-150) used in an ipx
or ss2? they're the ones I have access to at the
moment, no ipc's.

I've actually found an old sun bug report (1055000)
that mentions the problem specifically on the 4/75
(ss2) as being a testing bit flipped by the
manufacturer, and not being reset properly by the
kick-start routine, and ... no resolution was
offered...

I wonder if we used an ipc to initially set the chip,
whether that would fix the issue?

Like you, I've tried several commercial sources and
have gotten the same chips from both - apparently
sgs/thompson is the buyer/successor to mostek.

I haven't tried any Dallas Semiconductor chips, mainly
since the nvram faq said they mostly don't work
either...

So I will also try the m48t12-150 - thanks for the
reply!!

___________________________________________________________________
---- Harry J. Regan
---- Phone: [phone]
---- FAX: [phone]
---- ***@usa.net
---- ***@lepus.com


____________________________________________________________________
Get free email and a permanent address at http://www.amexmail.com/?A=1

[r***@sunhelp.org]

> > Then I ordered the 150 NS version of the M48T12 from JDR Microdevices (they
> > don't carry the M48T02), and it worked like a champ. They're $16.99 each from
> > JDR, rather than $12.80 from Mouser for the M48T02. Mouser also carries the
> > M48T12 chip, also at $12.80, but I haven't tried it (yet).
> >
> > So, despite what the NVRAM FAQ says, it looks like the MT48T12 chip is the one
> > to use. All the chips I've been playing with are from the same manufacturer,
> > so it's not a vendor problem.
> >
> > The ones that did NOT Work:
> >
> > Manufacturer: ST Label: M48T02-70PC1
> > Manufacturer: ST Label: M48T02-200PC1
> >
> > The one that DID Work:
> >
> > Manufacturer: ST Label: M48T12-150PC1
> >
> > I will sware that I successfully used an M48T02 on an IPC a while back.
> > Anyone out there have an IPC with a dead NVRAM chip willing to experiment with
> > one of these chips? I'll send a free chip to the first one who volunteers!


Just reporting back ...

I recently succeeded in rehabilitating an IPC using an M48T02-200PC1 from
Mouser, following instructions in the NVRAM FAQ.

I was going to try to transplant the chip into a Sparc-2, at a suggestion from
another list member. Our Sparc-2 proved to have a working NVRAM and I didn't
want to disturb its peace of mind.

-- SP

[Bran Tregare]

> > > I will sware that I successfully used an M48T02 on an IPC a while back.
> > > Anyone out there have an IPC with a dead NVRAM chip willing to
> experiment with
> > > one of these chips? I'll send a free chip to the first one who
> volunteers!
>
>
>Just reporting back ...
>
>I recently succeeded in rehabilitating an IPC using an M48T02-200PC1 from
>Mouser, following instructions in the NVRAM FAQ.


I just performed that operation using a M48T02-150PCI on one of my IPC's
and it seems happy.

[Eric Hall]

Bran Tregare wrote:

> > > > I will sware that I successfully used an M48T02 on an IPC a while back.
> > > > Anyone out there have an IPC with a dead NVRAM chip willing to
> > experiment with
> > > > one of these chips? I'll send a free chip to the first one who
> > volunteers!
> >

My NVRAM in my IPX just died!
Is there somplace I can order a
M48T12-150PC1 online?

Thank you, Eric H


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

[Harry Regan]

Just reporting back, though a couple of other folks have
said the same thing... The M48T02 NVRAM chips work like a
champ in IPCs-- no TOD clock problems. I got a couple of
IPCs in, swapped in the M48T02 chip and bingo! Nirvana!
Well, maybe not Nirvana, but the machine worked fine...

--- original message ---

Date: 3 Jan 01 08:17:18 EST
From: Harry Regan <***@usa.net>
To: ***@sunhelp.org
Subject: Re: [SunRescue] Bad - new NVRAM/TOD??
Reply-To: ***@sunhelp.org

Hi Steve--

Yeah, I experienced the same thing and posted a message about
it in December. Since then, I've found out two additional data
points. I originally ordered 200 NS chip for my IPX from Mouser
and got the TOD error you reference. I ordered a 70 NS version
of the same chip (also from Mouser) and got the same results.

Then I ordered the 150 NS version of the M48T12 from JDR
Microdevices (they don't carry the M48T02), and it worked like
a champ. They're $16.99 each from JDR, rather than $12.80 from
Mouser for the M48T02. Mouser also carries the M48T12 chip, also
at $12.80, but I haven't tried it (yet).

So, despite what the NVRAM FAQ says, it looks like the MT48T12
chip is the one to use. All the chips I've been playing with are
from the same manufacturer, so it's not a vendor problem.

The ones that did NOT Work:

Manufacturer: ST Label: M48T02-70PC1
Manufacturer: ST Label: M48T02-200PC1

The one that DID Work:

Manufacturer: ST Label: M48T12-150PC1

I will sware that I successfully used an M48T02 on an IPC a while
back. Anyone out there have an IPC with a dead NVRAM chip willing
to experiment with one of these chips? I'll send a free chip to the
first one who volunteers!

P.S. I have not found a way around the TOD Clock error. I even
tried some kinky nvramrc gymnastics, but the machine would always
blow back to the prom prompt.


___________________________________________________________________
---- Harry J. Regan
---- Phone: [phone]
---- FAX: [phone]
---- ***@usa.net
---- ***@lepus.com


____________________________________________________________________
Get free email and a permanent address at http://www.amexmail.com/?A=1

[Jeff Workman]

This post might be inappropriate. Click to display it.