Re: PPP fix for Sprint wireless network

[Matt Jerdonek <maj1224 at yahoo dot com>]

Hi Andrew,

No, I hadn't seen that patch before.  Thanks for
pointing it out to me.

The eCos PPP code is correct as it is today as I
interpret RFC 1662.  The Sprint network is definitely
where the problem is.  However, I didn't pursue
getting Sprint to fix the bug, because I didn't think
I'd be able to get a timely resolution.

When I was investigating this issue, I found that MS
Windows PCs are able to connect into Sprint without an
issue.  After analyzing the traces, I found that
Windows always sends the leading flag while eCos does
not.  After changing eCos, I was able to connect.  

I don't think we'd introduce any issues by sending
back-to-back flags.  According to RFC 1662:
"Two consecutive Flag Sequences constitute an empty
frame, which is silently discarded, and not counted as
an FCS error."  (Section 3.1)

In any case, I won't be upset if you choose to not
commit this patch.  I was just trying to save someone
else the time it took me to figure this out.

-- Matt

--- Andrew Lunn <andrew@lunn.ch> wrote:
> 
> Hi Matt
> 
> Have you seen:
> 
>
http://sources.redhat.com/ml/ecos-patches/2004-06/msg00029.html
> 
> A patch similar to your has been rejected before.
> Before accepting
> this i would like an greater understanding what is
> going on. Is sprint
> deliberately breaking the RFC, is it just a bug in
> Sprint's
> implementation? Or has there been a newer RFC which
> makes this flag a
> requirement?
> 
>         Thanks
>                 Andrew  
> 



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250