Michael Doerner wrote:
>
> Hi,
>
> I am running SuSE 6.2, kernel updated to 2.2.15, diald 0.99.4, external 56K
> modem.
>
> This was running very well for months (starting with diald 0.16.5) but since
> a few weeks, the modem dialup is not working reliable. A typical
> unsuccessful attempt looks like this :
>
> linux01 pppd[4537]: LCP: timeout sending Config-Requests
>
> With debugging on:
>
> May 31 18:45:39 linux01 pppd[748]: pppd 2.3.8 started by root, uid 0
> May 31 18:45:39 linux01 pppd[748]: Using interface ppp0
> May 31 18:45:39 linux01 pppd[748]: Connect: ppp0 <--> /dev/ttyS1
> May 31 18:45:39 linux01 pppd[748]: sent [LCP ConfReq id=0x1 <mru 1524>
> <asyncmap 0x0> <magic 0x13b31d10> <pcomp> <accomp>]
> May 31 18:46:06 linux01 last message repeated 9 times
> May 31 18:46:09 linux01 pppd[748]: LCP: timeout sending Config-Requests
>
> but the following redial attempt might then be successful like this:
>
> May 31 18:46:48 linux01 pppd[759]: Using interface ppp0
> May 31 18:46:48 linux01 pppd[759]: Connect: ppp0 <--> /dev/ttyS1
> May 31 18:46:48 linux01 pppd[759]: sent [LCP ConfReq id=0x1 <mru 1524>
> <asyncmap 0x0> <magic 0x84e2ac2a> <pcomp> <accomp>]
> May 31 18:46:51 linux01 pppd[759]: sent [LCP ConfReq id=0x1 <mru 1524>
> <asyncmap 0x0> <magic 0x84e2ac2a> <pcomp> <accomp>]
> May 31 18:46:51 linux01 pppd[759]: rcvd [LCP ConfAck id=0x1 <mru 1524>
> <asyncmap 0x0> <magic 0x84e2ac2a> <pcomp> <accomp>]
> May 31 18:46:52 linux01 pppd[759]: rcvd [LCP ConfReq id=0x0 <asyncmap 0x0>
> <auth pap> <magic 0x3beacf> <pcomp> <accomp>]
> May 31 18:46:52 linux01 pppd[759]: sent [LCP ConfAck id=0x0 <asyncmap 0x0>
> <auth pap> <magic 0x3beacf> <pcomp> <accomp>]
> May 31 18:46:52 linux01 pppd[759]: sent [PAP AuthReq id=0x1 user="xxxxxx"
> password=<hidden>]
> May 31 18:46:55 linux01 pppd[759]: sent [PAP AuthReq id=0x2 user="xxxxxx"
> password=<hidden>]
> May 31 18:46:55 linux01 pppd[759]: rcvd [PAP AuthAck id=0x2 ""]
> May 31 18:46:55 linux01 pppd[759]: Remote message: ^F
> May 31 18:46:55 linux01 pppd[759]: sent [IPCP ConfReq id=0x1 <addr 0.0.0.0>]
>
> The last 2 attempts were just in a row, no definition changes or any daemon
> restarts in between. I wonder whether it's something at the ISP's site?
> (Apparently, they are not very knowledgeable with Linux/Unix at all.
> Officially it's not supported ... but it was working perfect for months).
> I have tried updating pppd 2.3.11 in the meantime, no difference. I also
> tried other changes like increasing the amount of LCP requests
> (lcp-max-configure 20), changing mru size, etc. to no avail.
>
> Is there anything that I can trace at that stage? As far as I understand, I
> can't use i.e. ethereal for debugging purposes before there is an
> established connection on ppp0?
>
> Any help would be much appreciated.
> I am stuck here.
>
> Kind Regards,
> Michael Doerner
>
i keep getting a similar problem with UK's BT SurfTime. it is almost as
if about 1/3 of their servers are "broken" in some way. the effect is
very visible when i try to connect after 6pm, when access becomes
unmetered.
before 6pm about 1/3 of the connection attempts fail. between 6pm and
6:30pm, roughly 1/2 fail. then, between 6:30pm and about 8pm, almost all
attempts fail.
this looks like, out of a pool of 2/3 working and 1/3 "broken" servers,
the working ones slowly fill up with users, leaving only the broken ones
available during peak time. once the "peak" for "surfing" is past (9pm
to midnight), connection improves to 60% success again. this phenomenon
also seems to affect the service during weekends.
does anybody have any ideas?
cheers,
--
:D_ima
Dima Nemchenko
[EMAIL PROTECTED]
---------------------------------------------------------------------
"Eventually, every frog has to croak."
Louis, the "Budweiser Lizard"
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
