This is something we noticed a while ago. It was described to me at the
time by Livingston tech support as being correct. What they say is that a
lost connection due to something like pulling the line out of the back of
the user's modem while it's connected results in the telco issuing a
disconnect to the PM. It's as if the user at the other end of the phone
hung up the line. So, it's called a "User-Request" disconnect. I suppose
it's possible that just unplugging the channelized T1 is being considered
somehow the same. I'd sure prefer some other message though.
We were keeping track of the disconnect reasons and I still get the
daily stats from our system that compiles that data. However, I now just
toss it because it's essentially useless. For example, yesterday's data
shows one of our remote POP's with channelized T1 having 985 User-Request
disconnects and only 1 Lost-Carrier, which is hard to belive for users in
a rural area.
So, clearly you can't assume that the "User-Request" disconnect reason
means that the user's system actually requested a disconnect. This got me
into trouble with a customer real quick and I've ignored that data ever
since.
Yes, it would be nice if they'd fix that.
Chuck
On Wed, 22 Jul 1998, Russ Taylor wrote:
> On 7/22/98 8:03 PM, Tim Moloney (moloney@netsrq.com) wrote
>
> >Last Friday I converted my dialin lines from channelized T1 to PRI. I
> >had 7 users online when I moved the cable from the channelized T1 wall
> >jack to the the PRI wall jack. In the detail file, the
> >Acct-Termination-Cause for these users was User-Request. This is
> >clearly wrong. I would have expected Lost-Carrier, Lost-Service, or
> >Lost-Cable.
> >
> >In the past when a customer says that they've been cut off and I see
> >the termination cause was User-Request, I figured that Windows95 did
> >something behind their back or the user made a mistake. Now I'm not
> >so sure.
> >
> >Has anyone else noticed termination causes being incorrectly
> >classified as User-Request?
>
> My advice would be open a ticket with support, and send a RFE for a fix
> for the problem. People have brought this problem up multiple times in
> the past, and no Livingston has ever even acknowledged that the problem
> exists, let alone let us know if it will be fixed.
>
> This is one of those problems that doesn't exist as far as LE is
> concerned -- oh well!
>
> --
> Russ Taylor (rtaylor@cmc.net, http://www.cmc.net/~rtaylor/)
> Chambers Multimedia Connection Help Desk
> "Remember, you can't spell obsequious without I. O. U." -- Patty
>
>
>
> -
> To unsubscribe, email 'majordomo@livingston.com' with
> 'unsubscribe portmaster-users' in the body of the message.
> Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>
>
-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-users' in the body of the message.
Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>