Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (ASCEND) 5.0Ai28 - anyone have success?



Will Pierce wrote:
> 
> I have tried and tried and nothing I do seems to work to laod 5.0Ai28 onto
> a max 4048 we've got.
> 
> I've downloaded the tik.m40 and tk.m40's from ftp.ascend.com (binary mode)
> several times now.  I've tloadcode'd probably two dozen times...  Maybe I'm
> thick, but I can load 5.0Ai13, 5.0Ap27, 5.0Ap33, 5.0Ap36, even 6.0b1 with
> no problem whatsoever.  The permissions on these files are all the same on
> my tftp server.  I've tloaded from a config with nothing in it but the IP
> address/netmask (on same LAN/subnet as tftp server.)  Nadda.
> 
> Here's what happens-
> Instead of loading it as a fat load (which I think it should be, since it's
> greater than 960k in size), it only loads 5.0Ai28 as a thin load.  This happens
> for both tk.m40 and tik.m40.  It finishes the load, and then I do an nvramclear
> and it reboots.  As it boots, it comes up with "CRC error!!!" and then puts me
> into "RESTRICTED MODE * * * You must re-run the last tloadcode command" and
> my mod-16 cards show as mod-8 cards.  It looks like it could only boot up into
> a minimal kernel.  I then try simply rebooting, no go- same deal "restricted mode" again.
> I try tloadcode'ing 5.0Ai28 again, no go.
> 
> I've tried upping to 5.0Ai28 from 5.0Ai13, 5.0Ap27, 5.0Ap33, and 5.0Ap36 all
> to no avail.
> 
> Could there be a mistake in the code that figures out whether it is a fat or
> thin load?  Can I force a fat load?  (-f to tloadcode is only for forcing
> a tloadcode when no "platform identifier" is found in the image).
> 
> I dunno- maybe I've been tloadcode'ing wrong for the past year, but somehow
> I am beginning to think the code is messed up on ascend's ftp site.  I
> hesitate to suggest this, but I ran out of options.
> 
> So, I decided to load 6.0b1, and aside from a "Can't read debug profile" in
> my syslog, it looks okay.
> 
> I hope I'm being really stupid, and overlooked something in the docs.  Has
> anyone gotten 5.0Ai28 going?  
> 
> Thanks all....

Will- I've never loaded an incremental, just the "p" releases, but
I've gone through most of 'em from 4.6Ap12 to 5.0Ap33 and I've NEVER
seen it do a fat load. It's always been a thin load and it's always
worked fine.

I have been dying for user's download speeds since
> July of this year, when it was in 5.0Ai13, but rockwell code revisions have only
> been updated in the 'p' branch, so this is why I am so anxious to get 5.0Ai28
> working.

I'm not sure I understand this as it seems contradictory, but I assume
you meant "p" instead of "i" or vice-versa somewhere in there. At
any rate, Ap33 does have the updated Rockwell code. 

Also, I've never done an "nvram clear" on mine after updating the code-
I just reboot it. Basically, after getting the file into /tftpboot, I
just fsave, tsave to a null file I've created in /tftpboot, tloadcode,
(which does an fsave anyway) and when it loads with no errors (usually
does) I just reboot.

For all I know I'm not doing it right either, but it does seem to work
and they've all been thinloads. I've wondered about that too, since
some of 'em are bigger than 960K as you said- but I rather imagine
that THAT is caused by something silly, like the system reporting a
thinload no matter what it's really doing, or Ascend's docs being
a little out of sync. I've just never worried about it.

I went to bed at 4am and just woke up, so use the above at your own
risk. ;-) I may have missed something but I don't think so...

'Shot
++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>


References: