From: John Agosta on

"Philippe" <elipsia2001(a)yahoo.fr> wrote in message
news:elipsia2001-85D1E8.12422929112008(a)news.orange.fr...
> Hi all,
>
> I have at home a Cisco 857W router connected to an ADSL line. Very
> briefly, I need this equipment because I use the time-range based ACL's
> to manage Internet use at home. This i why I bought it instead of using
> any Netgers or Linksys box.
>
>
> I am quite far from the DLSAM (local loop is 4.600 m long ...) and the
> attenuation of the loop (52 dB) is very close to the upper limit (53
> dB).
>
> Unfortunately, I'm having a problem with my cisco box.
>
> When the cisco router is connected to the ADSL line, the attenuation
> raises to 56 dB (measured by the Cisco itself AND bye the DSLAM). When
> my ISP technician came home with its equipement and when its equipement
> is connected at exactly the same place than my router (just unplug the
> cable fom the router and plug it in the measuring equipement), the line
> attenuation is only 52 dB, confirmed by the DSLAM.
>
> I have asked the company that sold me the 857W for a new one, as it
> seems to me that 4 dB of loss is rather high.
> Unfortnualtely, they seems to be reluctant to replace this equipment,
> and are only sending me newer firmware "just in case it solves the
> problem".
>
> My question is : can anyone tell me wether 4 dB of loss is "normal" for
> a Cisco router connected to an ADSL line ? Do I have to insist until
> they send me another equipement ?
>
> Reagrds,
>
> PhC.



See if there is a "dsl gain-setting" command.
Don't know if it would help, but worth playing around with it, I guess.....





From: bod43 on
On 29 Nov, 17:58, alexd <troffa...(a)hotmail.com> wrote:
> Philippe wrote:
> > I have at home a Cisco 857W router connected to an ADSL line. Very
> > briefly, I need this equipment because I use the time-range based ACL's
> > to manage Internet use at home. This i why I bought it instead of using
> > any Netgers or Linksys box.
>
> Zyxel routers can do time-based access rules.
>
> > I am quite far from the DLSAM (local loop is 4.600 m long ...) and the
> > attenuation of the loop (52 dB) is very close to the upper limit (53
> > dB).
>
> I have heard many people recommend 2Wire ADSL routers for long lines, being
> able to hold onto sync with very low noise margins. Don't know if they do
> time-based ACLs, however.

Out of interest which country are you in
and who is your ISP? Getting a DSL
provider to even look at technical details
has been in my experience rather difficult.
(OK impossible).

I can confirm that there seem to be issues
with older 857 ADSL firmware. My experience has been
that some lines would not come up or were unreliable
until the firmware was changed. This was in the
UK with BT.

Definately worth trying an upgrade.

The other thing that I found to "work" on at
at least one ocassion was to change the
"dsl operating-mode" from auto to
the other options in turn. In the UK BT
for example works with either of the
options, ansi-dmt itu-dmt).
Of course of you are on ADSL2 then
these oprtions won't be suitable.
Easy to try.

Just so you know what is available:-
I am using IOS 12.4(15)T7 which comes with
DSL firmware "FW Version: 3.0.14".

sh dsl int atm0

The firmware is available seperately from
the IOS and is a free download.
Details on this newsgroup.

ftp://ftp.cisco.com/pub/access/800/
This seems to work again now but
has been 'resting' for a while.


You just put the file in the routers flash,
NOT accepting the "ERASE FLASH" option
if offered.
then rename it to whatever it says in the
release notes and reboot.

adsl_alc_20190.bin or something like that.

Note that the site (or maybe my firewall?)
requires passive ftp.

Site has also 3.0.33 and 4.10.9.
Read the release notes to make sure they
apply to your router. It may say that
it is only compatible with IOS x.x.x
but you basically need to ignore that
if you have a later IOS.

The ones that gave trouble were early 2.x.x.
I have had no problems at all with 3.0.x
but have not tried 4.x.x.

From: Philippe on
In article <elipsia2001-85D1E8.12422929112008(a)news.orange.fr>,
Philippe <elipsia2001(a)yahoo.fr> wrote:

Hi all,

thanks for all that.

I didn't know that Zyxel routers do have a "time range based ACL"
mechanism. it's maybe worth trying.

I'm in France, and my provider is Orange. I'm running IOS 12.4(15)T4,
wich seems to be the latest avalaible for my platform, according to
informations I found on Cisco docs.
My DSL firmware is 3.0.014; It's the release embedded with my IOS. I'v
tried with 3.0.33, 3.0.43 and 4.0.15, but none of them do change
anything to my problem (they seems to fix training problems with various
DSLAM's).

I will try to play a bit with the dsl gain-seting rx-offset command, but
I'm affraid it does't change anything. I'll keep you informed.

Regards,

PhC.
From: bod43 on
On 30 Nov, 08:59, Philippe <elipsia2...(a)yahoo.fr> wrote:
> In article <elipsia2001-85D1E8.12422929112...(a)news.orange.fr>,
>
>  Philippe <elipsia2...(a)yahoo.fr> wrote:
>
> Hi all,
>
> thanks for all that.
>
> I didn't know that Zyxel routers do have a "time range based ACL"
> mechanism. it's maybe worth trying.
>
> I'm in France, and my provider is Orange. I'm running IOS 12.4(15)T4,
> wich seems to be the latest avalaible for my platform, according to
> informations I found on Cisco docs.
> My DSL firmware is 3.0.014; It's the release embedded with my IOS. I'v
> tried with 3.0.33, 3.0.43 and 4.0.15, but none of them do change
> anything to my problem (they seems to fix training problems with various
> DSLAM's).
>
> I will try to play a bit with the dsl gain-seting rx-offset command, but
> I'm affraid it does't change anything. I'll keep you informed.

Thanks for the info regarding ISP.

Subsequent to my earlier post I noticed that the 4.x.x
firmware was not applicable to the 8[57]7.

This along with your statements lead me to wonder
if you actually managed to load the firmware
since you do not mention that 4.0.x failed.
You NEED to check with "sh dsl int atm 0".

"
......
Init FW: init_AMR-3.0.014_no_bist.bin
Operation FW: AMR-3.0.014.bin
FW Source: embedded
FW Version: 3.0.14
......"

This is what I see with the firmware integral to
my IOS.
If you succeed in loading the non integral firmware
that output looks quite different.


I have not personally had any problems at all with
firmware 3.x.x or later, only with some early 2.x.x.

One other thing worth looking at is to disconnect all
other phone equipment on the premises. In the UK
BT have a special diagnostic connection on the
"Master socket" that connects directly to the
outside - disconnecting ALL of the internal phone
connections. Maybe you have something similar.

Finally - and sadly - over a quite a few 10s of DSL
installations I came to the idea that there may
well be something sub-standard (not sub-Standard
necessarily) about the 8[57]7 routers. It
did seem to be the case that cheap basic routers
may well have worked better than the not so cheap
and not so basic Cisco stuff. On most installations
everything went well however there did seem to be
quite a few reports (I was rarely on-site myself) that
Brand-X is working and Cisco is not. We always
managed eventually to get the Cisco to work, but
the idea that there might be a deficiency became
quite well formed.

I previously did a lot of 837s (several 10s) and did
not notice the same effect with them.

Extra-finally:), you did not mention
"dsl operating-mode" . I have found that
fiddling with it was beneficial in some cases.

Good luck, please keep us in the loop.
From: Philippe on
In article <elipsia2001-917C5E.07262901122008(a)news.orange.fr>,
Philippe <elipsia2001(a)yahoo.fr> wrote:

> In article
> <63474e00-a415-45e0-8a53-e9c43e1c5e3e(a)3g2000yqs.googlegroups.com>,
> bod43 <Bod43(a)hotmail.co.uk> wrote:
>
> > You NEED to check with "sh dsl int atm 0".

Firmware before update :
Init FW: init_AMR-3.0.014_no_bist.bin
Operation FW: AMR-3.0.014.bin
FW Source: embedded
FW Version: 3.0.14

Firmware after update and reload :
Init FW: init_AMR-4.0.015.bin
Operation FW: AMR-4.0.015.bin
FW Source: external
FW Version: 4.0.15

So, it works :)

Regards,

Philippe.