Hi,

I have (had) a smoothly running ultra5 running Sol10-FCS.  I wanted to 
get it to run U1 (1/06), but the system would hang duing boot up - after 
the "Use is subject to terms" line.  The dial would spin for a bit, then 
just stop.  This would happen whether I used a CD or JS.

So, I tried to use the Update GUI.  After applying a few patches, and 
rebooting as appropriate, I hit a snag with one of the patches (I don't 
know which one, but I can have it narrowed if only I can boot it 
somehow).  Now I have no FCS CD's (they're at work), and I *just* 
deleted that release from my JS server due to space constraints.  :-(

If only I can boot it somehow, I can probably use 'patchrm -R' to back out.

Question - anyone know what is causing the ultra5 to hang?  And, anyone 
have another idea how I can recover tonight?

TIA,

Steve

Steven C. Liu wrote:

> Hi,
>
> I have (had) a smoothly running ultra5 running Sol10-FCS.  I wanted to 
> get it to run U1 (1/06), but the system would hang duing boot up - 
> after the "Use is subject to terms" line.  The dial would spin for a 
> bit, then just stop.  This would happen whether I used a CD or JS.
>
> So, I tried to use the Update GUI.  After applying a few patches, and 
> rebooting as appropriate, I hit a snag with one of the patches (I 
> don't know which one, but I can have it narrowed if only I can boot it 
> somehow).  Now I have no FCS CD's (they're at work), and I *just* 
> deleted that release from my JS server due to space constraints.  :-(
>
> If only I can boot it somehow, I can probably use 'patchrm -R' to back 
> out.
>
> Question - anyone know what is causing the ultra5 to hang?  And, 
> anyone have another idea how I can recover tonight?
>
> TIA,
>
> Steve

I've narrowed the problem down to 118822-27.  Unfortunately, in my 
attempts to back out, I've caused it to trash the backout script (?) 
*and* it now won't let me boot multi-user.  I get 
"/lib/svc/method/boot-archive" failed messages and the system wants me 
to repair it.  It's getting VERY late.  Sigh...

I guess it's time to rebuild the stinkin' system?  How does one recover 
the ability to back out KP 118822-27?  I need some serious help here 
folks...

BTW - the reason 118822-27 was a problem in the first place was because 
I have an USB ms/kb attached to one of the USB ports in a 3rd-party COTS 
Belkin PCI USB 2.0 card.  (The Belkin used to work just w/ my USB 
kb/ms.  I also have a LaCie drive attached to the ultra5 via this 
Belkin, and that is fine.)

My /etc/system used to say something like:
---
* Set the kb/ms to use usb
set consconfig:usb_kb_path="/pci@1f,0/pci@1/usb@2,1/device@1/keyboard@0"
set consconfig:usb_ms_path="/pci@1f,0/pci@1/usb@2,1/device@1/mouse@1"
---

Steve

Steven C. Liu wrote:

> Steven C. Liu wrote:
>
>> Hi,
>>
>> I have (had) a smoothly running ultra5 running Sol10-FCS.  I wanted 
>> to get it to run U1 (1/06), but the system would hang duing boot up - 
>> after the "Use is subject to terms" line.  The dial would spin for a 
>> bit, then just stop.  This would happen whether I used a CD or JS.
>>
>> So, I tried to use the Update GUI.  After applying a few patches, and 
>> rebooting as appropriate, I hit a snag with one of the patches (I 
>> don't know which one, but I can have it narrowed if only I can boot 
>> it somehow).  Now I have no FCS CD's (they're at work), and I *just* 
>> deleted that release from my JS server due to space constraints.  :-(
>>
>> If only I can boot it somehow, I can probably use 'patchrm -R' to 
>> back out.
>>
>> Question - anyone know what is causing the ultra5 to hang?  And, 
>> anyone have another idea how I can recover tonight?
>>
>> TIA,
>>
>> Steve
>
>
> I've narrowed the problem down to 118822-27.  Unfortunately, in my 
> attempts to back out, I've caused it to trash the backout script (?) 
> *and* it now won't let me boot multi-user.  I get 
> "/lib/svc/method/boot-archive" failed messages and the system wants me 
> to repair it.  It's getting VERY late.  Sigh...
>
> I guess it's time to rebuild the stinkin' system?  How does one 
> recover the ability to back out KP 118822-27?  I need some serious 
> help here folks...
>
> BTW - the reason 118822-27 was a problem in the first place was 
> because I have an USB ms/kb attached to one of the USB ports in a 
> 3rd-party COTS Belkin PCI USB 2.0 card.  (The Belkin used to work just 
> w/ my USB kb/ms.  I also have a LaCie drive attached to the ultra5 via 
> this Belkin, and that is fine.)
>
> My /etc/system used to say something like:
> ---
> * Set the kb/ms to use usb
> set consconfig:usb_kb_path="/pci@1f,0/pci@1/usb@2,1/device@1/keyboard@0"
> set consconfig:usb_ms_path="/pci@1f,0/pci@1/usb@2,1/device@1/mouse@1"
> ---
>
> Steve

Ok, folks, must get ready for work...  But, here's a last bit of 
additional information:

1. The system apparently hangs on the usb ms, not the kb.  Yes, I did a 
reconfigure, per the instruction in the cache.

My system uses logging, so I must have a way of reverting back to a 
specific date/time - right?  Please tell me how.  Please try to be as 
exact as possible, since I have now little room for error...

If I decide to move forward, how shall I configure my system to use a 
usb kb+ms?  Do I need to plunk in an USB 1.0 adaptor?  (I have one from 
Adaptec that is also certified by Sun.)  Do I need to change any module 
names in /etc/system?

TIA,

Steve

Steven C. Liu wrote:
> Steven C. Liu wrote:
> 
>> Steven C. Liu wrote:
>>

>>>
>>> If only I can boot it somehow, I can probably use 'patchrm -R' to 
>>> back out.
>>>

Your Jumpstart server will boot it:

"boot net" or "boot net:dhcp" (whichever way you've been installing)

Just leave off the "... - install" portion.

It's no different then booting from cd other than it's a sparse netimage.
You could try fixing your /etc/system first (from your comments below) 
then,
if that doesn't do it try booting and patchrm:

mount the appropriate disk slice under a temp mountpoint "/a"
and patchrm -R /a <patch-id>

and/or

vi /a/etc/system


>>
>> I guess it's time to rebuild the stinkin' system?  How does one 
>> recover the ability to back out KP 118822-27?  I need some serious 
>> help here folks...

See above.  Check the latest patch-release info or recommend bundle for 
the KUP or other patch-id's and see if it/they may fix your issue.  If 
it does, try patching your jumpstart image.  If it needs a third-party 
package then add that install to a finish script.

>>
>> BTW - the reason 118822-27 was a problem in the first place was 
>> because I have an USB ms/kb attached to one of the USB ports in a 
>> 3rd-party COTS Belkin PCI USB 2.0 card.  (The Belkin used to work just 
>> w/ my USB kb/ms.  I also have a LaCie drive attached to the ultra5 via 
>> this Belkin, and that is fine.)
>>
>> My /etc/system used to say something like:
>> ---
>> * Set the kb/ms to use usb
>> set consconfig:usb_kb_path="/pci@1f,0/pci@1/usb@2,1/device@1/keyboard@0"
>> set consconfig:usb_ms_path="/pci@1f,0/pci@1/usb@2,1/device@1/mouse@1"
>> ---
>>
>> Steve

Dunno anything about that, personally, but you can edit your system file 
and try a reboot to see if it works, ymmv.

> 
> 
> Ok, folks, must get ready for work...  But, here's a last bit of 
> additional information:
> 
> 1. The system apparently hangs on the usb ms, not the kb.  Yes, I did a 
> reconfigure, per the instruction in the cache.
> 
> My system uses logging, so I must have a way of reverting back to a 
> specific date/time - right?  Please tell me how.  Please try to be as 
> exact as possible, since I have now little room for error...
> 
> If I decide to move forward, how shall I configure my system to use a 
> usb kb+ms?  Do I need to plunk in an USB 1.0 adaptor?  (I have one from 
> Adaptec that is also certified by Sun.)  Do I need to change any module 
> names in /etc/system?

Like I said, dunno about that, sorry.

> 
> TIA,
> 
> Steve

HIH.