[c-nsp] boot images for 72xx to fix in 4M flash

[Rodney Dunn]

On Wed, Dec 14, 2005 at 12:42:44PM -0800, Ted Mittelstaedt wrote:
> 
> How exactly is a suggested fix that can only be done
> on the newer NPEs which have 8MB bootflash going
> to help the older NPE's with older rommons that have
> the 4MB bootflash? ;-)

What I meant was we are going to try and get a later 
12.0S boot image small enough to fit in a 4M
bootflash. If we can't we'll leave the older two out there.

I don't know the answer to your other questions.



> 
> Seriously, though, how many devices does this affect?
> 
> Just out of curiosity I went to our spares and pulled
> the FE I/O cards out of the 7206's we have in spares,
> both of these appear to have 1 SOCKETED flash SIMM.
> (they both come up as 4MB bootflash devices)
> Aren't these the bootflash devices?  Can't these be
> replaced with larger flash SIMMS?
> 
> Or am I looking at the wrong thing?
> 
> I also noticed the rommon chips are EPROMS,
> both of them are labeled ROMMON V11.1(10)
> Is it possible to download binary images of later rommons
> from Cisco and use an eprom programmer to update
> these so that they will boot directly from the pcmcia
> flash cards?  Or can I order a new rom?
> 
> Ted
> 
> 
> >-----Original Message-----
> >From: Rodney Dunn [mailto:rodunn at cisco.com]
> >Sent: Tuesday, December 13, 2005 5:08 PM
> >To: Ted Mittelstaedt
> >Cc: Rodney Dunn; Ian Dickinson; Robert E.Seastrom;
> >cisco-nsp at puck.nether.net
> >Subject: Re: [c-nsp] boot images for 72xx to fix in 4M flash
> >
> >
> >It's better to just not load the boot image at all
> >if it's not required. The newer rommons on newer NPE's
> >can do that. There isn't any development going to be
> >done on these older boot images other than just
> >fixing a bug if one shows up. They *may* if possible
> >try and get a more recent image to fit by inspecting
> >yet again what has creeped in. It may not happen but
> >they are going to try. I don't expect they will do
> >anything above that at this stage in the product lifecycle.
> >
> >Rodney
> >
> >On Mon, Dec 12, 2005 at 10:36:34PM -0800, Ted Mittelstaedt wrote:
> >>
> >>
> >> >-----Original Message-----
> >> >From: Rodney Dunn [mailto:rodunn at cisco.com]
> >> >Sent: Monday, December 12, 2005 10:29 AM
> >> >To: Ted Mittelstaedt
> >> >Cc: Ian Dickinson; Rodney Dunn; Robert E.Seastrom;
> >> >cisco-nsp at puck.nether.net
> >> >Subject: Re: [c-nsp] boot images for 72xx to fix in 4M flash
> >> >
> >> >
> >> >On Mon, Dec 12, 2005 at 09:23:59AM -0800, Ted Mittelstaedt wrote:
> >> >>
> >> >> I would guess then that 12.0.23 S6 is deferred as well.
> >> >
> >> >They were deferred but without going in to the details
> >> >I was able to get them posted back. I'll have to leave it
> >> >at that.
> >> >
> >>
> >> This came up once before, as I recall Cisco deferred a bunch of
> >> boot images for exceeding the 4MB size a couple years ago.
> >>
> >> Obviously someone forgot about doing that.
> >>
> >> >
> >> >We are going to try our best to get the image size down if
> >> >possible via:
> >> >
> >> >CSCsc63112
> >> >Externally found moderate defect: Assigned (A)
> >> >7200 boot images exceed 4M bootflash due to image size growth
> >> >
> >> >*without* removing old device driver support but it may be
> >> >impossible but at least they are going to try.
> >> >
> >>
> >> Rodney,
> >>
> >> Would it be possible to post TWO boot images of the same version,
> >> one that has ONLY support for ethernet PA's and the other which
> >> has ONLY support for serial PAs?
> >>
> >> The idea would be that you could then fit all of the variations of the
> >> ethernet PA's in one image, most people would use that, and all
> >> the variations of serial PA's in the other image which would be used
> >> by folks remote-booting 7200 series?
> >>
> >> One advantage of that as well would be for example when we reboot
> >> our 7206vxr (with a 4MB bootflash chip) when it comes up the bootcode
> >> helpfully activates all the atm and serial interfaces, then those all
> >> almost
> >> immediately deactivate when the main image is loading, then reactivate
> >> again.  As several of those interfaces go to feeds, the BGP
> >peers on the
> >> other end see us come up, then go down, then come up again.  I'd just
> >> as soon not activate those interfaces during the loading of the boot
> >> code,
> >> and removing support for the ATM PA"s from the bootflash would thus
> >> have a helpful side-effect.
> >>
> >> >
> >> >> Netflow is probably completely useless either way for this.
> >> >
> >> >It had nothing to do with this for the boot images it was
> >> >done for consistency which is the only reason I was able
> >> >to get those two posted back for now until and if we can
> >> >get it really resolved via: CSCsc63112
> >> >
> >>
> >> I figured that, but I couldn't resist poking fun anyhow.
> >>
> >> Ted
> >
> >--
> >No virus found in this incoming message.
> >Checked by AVG Free Edition.
> >Version: 7.1.371 / Virus Database: 267.13.13/199 - Release
> >Date: 12/13/2005
> >