US Robotics SureWontConnect 9003 ADSL Router

Not sure if anyone will remember the thread(s) here and on uk.broadband
about the USR SureConnect Router
but thought you might enjoy hearing the ongoing saga of my experiences. I
did say I'd let you know how I got on.
Basically it looks like a timing issue with their PPP/IPCP stack thats
become more or an issue after something got
changed either by Pipex or BT.

After realising USRs PPP stack was non complient ( or just bust ) I tried
getting USR to fix it and then attempted
to return it as not fit for use to Ebuyer where I bought it. One of these
actions I reckoned might allow me to get
a working ADSL Router, alas as is so often the case its not that simple.

1) Ebuyer refused to refund or replace with another make of router. They did
send me a brand spanking new non
working router unlike the one they originally sent which has someone elses
config in it.

I did some more test with it just to prove my sanity and managed during
repeated connection attempts to collect
the following exceptions.

ppp: LCP : ppp0: LayerUp
Address store Exception
Exception Program Counter: 0x8028d3f8
Status Register: 0x3000ff01
Cause Register: 0x00000014
Access Address : 0x10f4a1fa
Task: 0x80547b40 "tmrs"
pppatalink : link name ppp0, Authentication phase: mine = none

After about 30 attempts the ipct task dies requiring a reboot...
ppp: FSM: ppp0: State change Ack-Sent -->
Opened
ppp: LCP : ppp0: LayerUp
Address load Exception
Exception Program Counter: 0x8029ce98
Status Register: 0x3000ff00
Cause Register: 0x00000010
Access Address : 0x80c4742e
Task: 0x80fffde0 "ipct"

I now have a very slick sequence I could script to get a connection if it
wasnt
for the two tasks stopping.

1) Watch the connect get stuck at negotiating in the web Gui
2) telnet in and Enable ppptrace
3) select pppoastop 0
4) select pppoastart 0
5) Check the trace on the serial port and see if "Oops up at opened"
appears,
if so repeat steps 3-5.

Much like its predecessor I did get a working connection where
"ppp: FSM ppp0: Oops, Up at Opened" was avoided. Anyway I cant be bothered
manually attemting the connect 50 times when I want to log on so I'm back to
my
trusty Alcatel Modem.

2) USR ((E-Mail Removed)) initially did what I expected and quoted
the
manual at me. To this I made a fairly curt response and sent them the trace
AGAIN and pointed out the tell tale "Oops" comment in the trace.
I then suggested they got some kind of engineering type involved who knew
a) how to use grep and b)understood what a state machine was.

After some delay I got a "If your still having problems let us know" type
email.
To which I responded "yes your products still a bag of $hit please can you
fix it"

They then asked me to send my trace to a new email address
(E-Mail Removed) which initially
bounced. I've heard nothing from this lot yet.

If this ever works I'll let you all know.

nospam-(E-Mail Removed) wrote in
news:3f856a85$0$27819$(E-Mail Removed) :

> Not sure if anyone will remember the thread(s) here and on
> uk.broadband about the USR SureConnect Router
> but thought you might enjoy hearing the ongoing saga of my
> experiences. I did say I'd let you know how I got on.
> Basically it looks like a timing issue with their PPP/IPCP stack thats
> become more or an issue after something got
> changed either by Pipex or BT.
>
Can't say I understood all of your post but I have one of these devices on
Plusnet and it does work for basic ADSL. I've not tried owt to special but
have Linux and Windoze quite happy with all the bits I need except

The DNS relay is definitely yukk'd.

Not been on here for ages, no excuses, I know.
I've come on tonight, though, to ask if any one has managed to get the 9003
router to connect to Nildram.
Don't know if it has been mentioned before, but I can't get the 9003 router
to connect to Nildram. Web interface shows stuck on "negotiating". It works
fine on BT Broadband, but not my preferred isp Nildram.
I got the crappy robotic response from mail to usr support, so have also
sent the snotty message back. Not impressed.

Mike

<nospam-(E-Mail Removed)> wrote in message
news:3f856a85$0$27819$(E-Mail Removed) ...
> Not sure if anyone will remember the thread(s) here and on uk.broadband
> about the USR SureConnect Router
> but thought you might enjoy hearing the ongoing saga of my experiences. I
> did say I'd let you know how I got on.
> Basically it looks like a timing issue with their PPP/IPCP stack thats
> become more or an issue after something got
> changed either by Pipex or BT.
>
> After realising USRs PPP stack was non complient ( or just bust ) I tried
> getting USR to fix it and then attempted
> to return it as not fit for use to Ebuyer where I bought it. One of these
> actions I reckoned might allow me to get
> a working ADSL Router, alas as is so often the case its not that simple.
>
> 1) Ebuyer refused to refund or replace with another make of router. They
did
> send me a brand spanking new non
> working router unlike the one they originally sent which has someone elses
> config in it.
>
> I did some more test with it just to prove my sanity and managed during
> repeated connection attempts to collect
> the following exceptions.
>
> ppp: LCP : ppp0: LayerUp
> Address store Exception
> Exception Program Counter: 0x8028d3f8
> Status Register: 0x3000ff01
> Cause Register: 0x00000014
> Access Address : 0x10f4a1fa
> Task: 0x80547b40 "tmrs"
> pppatalink : link name ppp0, Authentication phase: mine = none
>
> After about 30 attempts the ipct task dies requiring a reboot...
> ppp: FSM: ppp0: State change Ack-Sent -->
> Opened
> ppp: LCP : ppp0: LayerUp
> Address load Exception
> Exception Program Counter: 0x8029ce98
> Status Register: 0x3000ff00
> Cause Register: 0x00000010
> Access Address : 0x80c4742e
> Task: 0x80fffde0 "ipct"
>
> I now have a very slick sequence I could script to get a connection if it
> wasnt
> for the two tasks stopping.
>
> 1) Watch the connect get stuck at negotiating in the web Gui
> 2) telnet in and Enable ppptrace
> 3) select pppoastop 0
> 4) select pppoastart 0
> 5) Check the trace on the serial port and see if "Oops up at opened"
> appears,
> if so repeat steps 3-5.
>
> Much like its predecessor I did get a working connection where
> "ppp: FSM ppp0: Oops, Up at Opened" was avoided. Anyway I cant be bothered
> manually attemting the connect 50 times when I want to log on so I'm back
to
> my
> trusty Alcatel Modem.
>
> 2) USR ((E-Mail Removed)) initially did what I expected and
quoted
> the
> manual at me. To this I made a fairly curt response and sent them the
trace
> AGAIN and pointed out the tell tale "Oops" comment in the trace.
> I then suggested they got some kind of engineering type involved who knew
> a) how to use grep and b)understood what a state machine was.
>
> After some delay I got a "If your still having problems let us know" type
> email.
> To which I responded "yes your products still a bag of $hit please can you
> fix it"
>
> They then asked me to send my trace to a new email address
> (E-Mail Removed) which initially
> bounced. I've heard nothing from this lot yet.
>
> If this ever works I'll let you all know.

Other than its a nightmare to connect it does shift packets OK. I'm not
doing anything fancy
with it, its just the connect thats the problem. Plus its only become a
problem in the last month
or so with Pipex. Which makes me suspect somethings changed. Regardless the
trace shows
its not working properly. The damn thing should connect or at least
re-attempt a connection under
its own steam.

I could of course just leave it connected all the time, but personally dont
like stuff powered up 24hrs
a day.

Dave

Sounds like the same problem, its a bust implementation of PPP negotiation.
Their finite state machine does not work properly.

I suspect if you follow my sequence below you will eventually get a
connection. Does the web Gui show
it has an IP address? Mine gets the address goes active then returns to
negotiating.

1) Watch the connect get stuck at negotiating in the web Gui
2) telnet in and Enable ppptrace (type "sndcp", then "ppptrace on")
3) select "pppoastop 0"
4) select "pppoastart 0"
5) Check the trace on the serial port and see if "Oops up at opened"
appears, if so repeat steps 3-5.

Alternatively pull out the plug and reinsert it. Note if you get either of
the exceptions below you need to power it
off/on before trying again.

Dave

nospam-(E-Mail Removed) wrote:

> Other than its a nightmare to connect it does shift packets OK. I'm not
> doing anything fancy
> with it, its just the connect thats the problem. Plus its only become a
> problem in the last month
> or so with Pipex. Which makes me suspect somethings changed. Regardless
> the trace shows
> its not working properly. The damn thing should connect or at least
> re-attempt a connection under
> its own steam.
>
> I could of course just leave it connected all the time, but personally
> dont like stuff powered up 24hrs
> a day.
>
> Dave

I saw a post where someone got round problems connecting by using pppoe, at
the time I thought you couldn't do that, but apparently BT have been
trialing it and have it sorted now. I haven't got a clue if it will work
with pipex - I may try just to see, but it's alot of hassle for me with a
PCI/Linux and is AFAIK a bit less efficient.

Could be worth a go if it's easy for you with a router - you may need to
set MTU to 1492 on clients.

Andy.