UMTS
in [Linux Networking]
|
From: Marc Haber on 20 Mar 2010 04:18 ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >On Wed, 17 Mar 2010, in the Usenet newsgroup comp.os.linux.networking, in >article <hnqdan$vg9$1(a)news1.tnib.de>, Marc Haber wrote: >>(Moe Trin) wrote: > >>>Which version of ppp are you using? > >>2.4.4rel-10.1 from Debian unstable, I have been seeing this behavior >>ever since I started using mobile Internet access years ago. > >You may want to check with the Debian maintainer, as the ppp package >supplied by Debian is substantially modified from the ANU source from >Mackerras and Carlson. The Debian package of ppp surely needs some love by its maintainers, agreed. > None-the-less, this does appear to be a >problem with the wireless/mobile/UMTS/3g system rather than in pppd. With all of them? I do have experience with multiple vendor's products. >>Since I have changed over to wvdial, my pppd doesn't log the IPCP/LCP >>conversations any more. I cannot verify this. > >You could by adding 'debug' to /etc/ppp/options, but wvdial being a >windoze wanna-be program tries to hide technical stuff like that. As long as I don't have to fight with chat and the painful wishes it leaves unsatisfied due to its inability to properly handle loops and/or errors, wvdial is just fine. It leaves some wishes, yes. > >>Since the PPP Peer in this case is the UMTS device and I have seen >>this behavior with a lot of different UTMS devices - is proposing >>10.11.12.13 and 10.11.12.14 an official standard, or just a >>convention among all vendors? > >I haven't found any official standards - and it appears to be a >rather unfortunate assumption by the vendors trying to choose a >number that users would _never_ think of using themselves. To bad >the idiots never asked a 6 or 7 year old child. I still think that it is highly unlikely that all vendors have come to the same "solution" independently. Greetings Marc -- -------------------------------------- !! No courtesy copies, please !! ----- Marc Haber | " Questions are the | Mailadresse im Header Mannheim, Germany | Beginning of Wisdom " | http://www.zugschlus.de/ Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834
From: Günther Schwarz on 20 Mar 2010 04:51 Marc Haber wrote: > ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >>I haven't found any official standards - and it appears to be a rather >>unfortunate assumption by the vendors trying to choose a number that >>users would _never_ think of using themselves. To bad the idiots never >>asked a 6 or 7 year old child. > > I still think that it is highly unlikely that all vendors have come to > the same "solution" independently. How many vendors are out there? Ericsson, Alcatel-Lucent, Huawei, Nokia- Siemens, some more? It might well be that you are talking to the very same implementation on competing 3G networks. Günther
From: Moe Trin on 20 Mar 2010 22:24 On Sat, 20 Mar 2010, in the Usenet newsgroup comp.os.linux.networking, in article <ho20c2$i5j$1(a)news1.tnib.de>, Marc Haber wrote: >ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >>You may want to check with the Debian maintainer, as the ppp package >>supplied by Debian is substantially modified from the ANU source from >>Mackerras and Carlson. >The Debian package of ppp surely needs some love by its maintainers, >agreed. Is it being maintained? Last I bothered to look at the diff file, it had a change adding 'auth' to the /etc/ppp/options file: ============ +# Require the peer to authenticate itself before allowing network +# packets to be sent or received. +# Please do not disable this setting. It is expected to be standard in +# future releases of pppd. Use the call option (see manpage) to disable +# authentication for specific peers. +auth ============ which probably came from a note added to the ChangeLog for ppp-2.3.0 in 1997. The actual authentication changes to Linux/Solaris/DGUX versions occurred in 2.3.6 over 11 years ago (see the Changes-2.3 in the ppp source), but never did make authentication mandatory for all situations as this addition does. See the 'auth' option and the rather lengthy 'AUTHENTICATION' section in the man page for details. >> None-the-less, this does appear to be a problem with the >> wireless/mobile/UMTS/3g system rather than in pppd. >With all of them? I do have experience with multiple vendor's >products. Can't say for sure, but I've seen reports for multiple vendors. I don't know how many different chipsets are actually involved. >> You could by adding 'debug' to /etc/ppp/options, but wvdial being a >> windoze wanna-be program tries to hide technical stuff like that. >As long as I don't have to fight with chat and the painful wishes it >leaves unsatisfied due to its inability to properly handle loops >and/or errors, wvdial is just fine. I can't say that I've never had a problem with chat (quoting escape characters in the command line), but it's been many years (~15) since I did have a problem. >It leaves some wishes, yes. WvDial is built on the wrong premise - it expects to be able to log in to a UNIX shell on the remote, and wants to find a Login: prompt to do so. That this mode of operation became rare when microsoft invented the telephone back in 1995 is apparently lost on the authors of wvdial. Their default modem init sequences are also non-standard. >>> is proposing 10.11.12.13 and 10.11.12.14 an official standard, or >>> just a convention among all vendors? >> I haven't found any official standards - and it appears to be a >> rather unfortunate assumption by the vendors trying to choose a >> number that users would _never_ think of using themselves. To bad >> the idiots never asked a 6 or 7 year old child. >I still think that it is highly unlikely that all vendors have come to >the same "solution" independently. A search through the RFCs doesn't turn up anything, and I'm left with the suspicion that it's one chip vendor's idea that may be an assumed standard rather than a documented one. Old guy
From: Marc Haber on 21 Mar 2010 08:51 Günther Schwarz <strap(a)gmx.de> wrote: >Marc Haber wrote: >> ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >>>I haven't found any official standards - and it appears to be a rather >>>unfortunate assumption by the vendors trying to choose a number that >>>users would _never_ think of using themselves. To bad the idiots never >>>asked a 6 or 7 year old child. >> >> I still think that it is highly unlikely that all vendors have come to >> the same "solution" independently. > >How many vendors are out there? Ericsson, Alcatel-Lucent, Huawei, Nokia- >Siemens, some more? Novatel, Option are the ones that immediately jump to my mind. > It might well be that you are talking to the very >same implementation on competing 3G networks. So you're talking equipment on the network side? That might be possible, yes. Greetings Marc -- -------------------------------------- !! No courtesy copies, please !! ----- Marc Haber | " Questions are the | Mailadresse im Header Mannheim, Germany | Beginning of Wisdom " | http://www.zugschlus.de/ Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834
From: Marc Haber on 21 Mar 2010 08:53
ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >On Sat, 20 Mar 2010, in the Usenet newsgroup comp.os.linux.networking, in >article <ho20c2$i5j$1(a)news1.tnib.de>, Marc Haber wrote: >>ibuprofin(a)painkiller.example.tld.invalid (Moe Trin) wrote: >>>You may want to check with the Debian maintainer, as the ppp package >>>supplied by Debian is substantially modified from the ANU source from >>>Mackerras and Carlson. > >>The Debian package of ppp surely needs some love by its maintainers, >>agreed. > >Is it being maintained? I think so. > Last I bothered to look at the diff file, it >had a change adding 'auth' to the /etc/ppp/options file: > >============ >+# Require the peer to authenticate itself before allowing network >+# packets to be sent or received. >+# Please do not disable this setting. It is expected to be standard in >+# future releases of pppd. Use the call option (see manpage) to disable >+# authentication for specific peers. >+auth >============ > >which probably came from a note added to the ChangeLog for ppp-2.3.0 >in 1997. Then one should file a bug in the Debian BTS to have the comment removed from the Debian package. Such explanations tend to outlive their purpose since one needs really intimate knowledge of the Package to be able to judge whether it still applies or not. >WvDial is built on the wrong premise - it expects to be able to log in >to a UNIX shell on the remote, and wants to find a Login: prompt to do >so. That this mode of operation became rare when microsoft invented the >telephone back in 1995 is apparently lost on the authors of wvdial. >Their default modem init sequences are also non-standard. Indeed. Stupid mode works fine though. It leaves some wishes as well, but they are not half as annoying as chat(8)'s shortcomings. Greetings Marc -- -------------------------------------- !! No courtesy copies, please !! ----- Marc Haber | " Questions are the | Mailadresse im Header Mannheim, Germany | Beginning of Wisdom " | http://www.zugschlus.de/ Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834 |