Using PPP to dial-in to ISP - Mandriva

This is a discussion on Using PPP to dial-in to ISP - Mandriva ; Moe Trin wrote: > On Fri, 22 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article > , Daniel wrote: > >> > Find a command prompt, and run the commands '/sbin/ifconfig -a' and > '/sbin/route -n' before, during and ...

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 21 to 39 of 39

Thread: Using PPP to dial-in to ISP

  1. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Fri, 22 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47bdf990$0$26072$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >>




    > Find a command prompt, and run the commands '/sbin/ifconfig -a' and
    > '/sbin/route -n' before, during and after dialing in.
    >


    Before dial-in:-
    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:73 errors:0 dropped:0 overruns:0 frame:0
    TX packets:73 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:5241 (5.1 KiB) TX bytes:5241 (5.1 KiB)

    sit0 Link encap:IPv6-in-IPv4
    NOARP MTU:1480 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    [daniel@localhost ~]$ /sbin/route -n
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use
    Iface
    169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0
    lo
    127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0
    lo


    During Dial-in:-
    [daniel@localhost ~]$ /sbin/ifconfig -a
    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:73 errors:0 dropped:0 overruns:0 frame:0
    TX packets:73 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:5241 (5.1 KiB) TX bytes:5241 (5.1 KiB)

    ppp0 Link encap:Point-to-Point Protocol
    POINTOPOINT NOARP MULTICAST MTU:1500 Metric:1
    RX packets:0 errors:1 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:3
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    sit0 Link encap:IPv6-in-IPv4
    NOARP MTU:1480 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    [daniel@localhost ~]$ /sbin/route -n
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use
    Iface
    202.3.39.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

    While connected:-
    [daniel@localhost ~]$ /sbin/ifconfig -a
    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:73 errors:0 dropped:0 overruns:0 frame:0
    TX packets:73 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:5241 (5.1 KiB) TX bytes:5241 (5.1 KiB)

    ppp0 Link encap:Point-to-Point Protocol
    inet addr:202.3.39.21 P-t-P:202.3.39.1 Mask:255.255.255.255
    UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
    RX packets:4 errors:1 dropped:0 overruns:0 frame:0
    TX packets:4 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:3
    RX bytes:64 (64.0 b) TX bytes:82 (82.0 b)

    sit0 Link encap:IPv6-in-IPv4
    NOARP MTU:1480 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    [daniel@localhost ~]$ /sbin/route -n
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use
    Iface
    202.3.39.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

    After Hanging Up:-
    [daniel@localhost ~]$ /sbin/ifconfig -a
    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:75 errors:0 dropped:0 overruns:0 frame:0
    TX packets:75 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:5341 (5.2 KiB) TX bytes:5341 (5.2 KiB)

    sit0 Link encap:IPv6-in-IPv4
    NOARP MTU:1480 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    [daniel@localhost ~]$ /sbin/route -n
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use
    Iface
    169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

    > What else is on the Ethernet link?
    >


    Nothing and the card has been removed and returned to ISP.

    >> What do I need to do to change this arrangement?? Can you answer here,
    >> or would you prefer I start a new thread??

    >
    > This should do it. One last check is to see that WHILE DIALED IN, there
    > are appropriate nameservers listed in /etc/resolv.conf.
    >
    > Old guy


    Missed this last one, will do it tomorrow, if you still think it will help.

    Again, thanks.

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  2. Re: Using PPP to dial-in to ISP

    On Fri, 22 Feb 2008 05:30:05 -0500, Daniel wrote:

    > While connected:-
    > [daniel@localhost ~]$ /sbin/route -n
    > Kernel IP routing table
    > Destination Gateway Genmask Flags Metric Ref Use
    > Iface
    > 202.3.39.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    > 169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    > 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo


    This one, I can help with. from my system, while connected ...
    # route -n
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use Iface
    216.240.2.252 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
    0.0.0.0 0.0.0.0 0.0.0.0 U 0 0 0 ppp0

    Note the last line. kppp should have added the defaultroute command line
    option, but apparently didn't. You can add "defaultroute" to
    /etc/ppp/options, to ensure it does get set. Technically the last line
    should have the Flags set to UG, but the above does work. Running route
    without the -n flag gives me
    # route
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use Iface
    ts2.ody.ca * 255.255.255.255 UH 0 0 0 ppp0
    127.0.0.0 * 255.0.0.0 U 0 0 0 lo
    default * 0.0.0.0 U 0 0 0 ppp0

    Regards, Dave Hodgins

    --
    Change nomail.afraid.org to ody.ca to reply by email.
    (nomail.afraid.org has been set up specifically for
    use in usenet. Feel free to use it yourself.)

  3. Re: Using PPP to dial-in to ISP

    On Fri, 22 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47be98ac$0$29827$88260bb3@free.teranews.com>, Daniel wrote:

    >Before dial-in:-


    >lo Link encap:Local Loopback


    >sit0 Link encap:IPv6-in-IPv4


    OK

    >[daniel@localhost ~]$ /sbin/route -n
    >Kernel IP routing table
    >Destination Gateway Genmask Flags Metric Ref Use Iface
    >169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    >127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo


    The first route (169.254.0.0) is ridiculous, but isn't hurting things.
    This is "ZeroConf" (RFC3927), and is the address range a system would
    use if it can't find a DHCP server. In 30 years of working with IP, I've
    never seen a DHCP server on the Loopback interface. You could disable
    this by inserting/setting a line in /etc/sysconfig/network that says

    NOZEROCONF=yes

    >During Dial-in:-
    >[daniel@localhost ~]$ /sbin/ifconfig -a
    >lo Link encap:Local Loopback


    OK

    >ppp0 Link encap:Point-to-Point Protocol
    > POINTOPOINT NOARP MULTICAST MTU:1500 Metric:1
    > RX packets:0 errors:1 dropped:0 overruns:0 frame:0
    > TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    > collisions:0 txqueuelen:3
    > RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)


    I would have expected to see addresses as below

    >sit0 Link encap:IPv6-in-IPv4


    OK

    >[daniel@localhost ~]$ /sbin/route -n
    >Kernel IP routing table
    >Destination Gateway Genmask Flags Metric Ref Use Iface
    >202.3.39.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    >169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    >127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo


    and there's the problem. This tells the kernel how to talk to the peer
    and to yourself, but not how to reach the Internet. There is a switch
    or tic-box in the KPPP configuration that tells kppp to make this the
    "default route". The other option might be to simply include the option
    "defaultroute" (yes, that's no space) in /etc/ppp/options or similar.

    defaultroute
    Add a default route to the system routing tables, using
    the peer as the gateway, when IPCP negotiation is success-
    fully completed. This entry is removed when the PPP con-
    nection is broken.

    >> What else is on the Ethernet link?

    >
    >Nothing and the card has been removed and returned to ISP.


    OK - one less thing to go wrong.

    >> This should do it. One last check is to see that WHILE DIALED IN, there
    >> are appropriate nameservers listed in /etc/resolv.conf.


    >Missed this last one, will do it tomorrow, if you still think it will help.


    After the default routing issue is fixed, this problem would show up as
    being able to connect to 202.158.214.106, but NOT be able to find
    mirror.aarnet.edu.au (ftp://mirror.aarnet.edu.au/pub/ibiblio/ is a mirror
    of the Sunsite.unc.edu or Ibiblio.org software repository that is only
    accessible in Oz).

    >Again, thanks.


    You're doing fine. Glad to be able to help.

    Old guy

  4. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Fri, 22 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47be98ac$0$29827$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> Before dial-in:-

    >
    >> lo Link encap:Local Loopback

    >
    >> sit0 Link encap:IPv6-in-IPv4

    >
    > OK
    >
    >> [daniel@localhost ~]$ /sbin/route -n
    >> Kernel IP routing table
    >> Destination Gateway Genmask Flags Metric Ref Use Iface
    >> 169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    >> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

    >
    > The first route (169.254.0.0) is ridiculous, but isn't hurting things.
    > This is "ZeroConf" (RFC3927), and is the address range a system would
    > use if it can't find a DHCP server. In 30 years of working with IP, I've
    > never seen a DHCP server on the Loopback interface. You could disable
    > this by inserting/setting a line in /etc/sysconfig/network that says
    >
    > NOZEROCONF=yes
    >
    >> During Dial-in:-
    >> [daniel@localhost ~]$ /sbin/ifconfig -a
    >> lo Link encap:Local Loopback

    >
    > OK
    >
    >> ppp0 Link encap:Point-to-Point Protocol
    >> POINTOPOINT NOARP MULTICAST MTU:1500 Metric:1
    >> RX packets:0 errors:1 dropped:0 overruns:0 frame:0
    >> TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    >> collisions:0 txqueuelen:3
    >> RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

    >
    > I would have expected to see addresses as below
    >
    >> sit0 Link encap:IPv6-in-IPv4

    >
    > OK
    >
    >> [daniel@localhost ~]$ /sbin/route -n
    >> Kernel IP routing table
    >> Destination Gateway Genmask Flags Metric Ref Use Iface
    >> 202.3.39.1 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
    >> 169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 lo
    >> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo

    >
    > and there's the problem. This tells the kernel how to talk to the peer
    > and to yourself, but not how to reach the Internet. There is a switch
    > or tic-box in the KPPP configuration that tells kppp to make this the
    > "default route". The other option might be to simply include the option
    > "defaultroute" (yes, that's no space) in /etc/ppp/options or similar.
    >


    A tick in the KPPP configuration box and here I am using MD2007 again!

    Thanks for all your assistance, Old Guy and David.

    Now, I guess I can go through the various files I've edited and see what
    I can remove to simplify the dial-in process.

    Then I can try switching from PAP to CHAP but that might be a little
    while off.

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  5. Re: Using PPP to dial-in to ISP

    On Sat, 23 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47bf58c3$0$25987$88260bb3@free.teranews.com>, Daniel wrote:

    >A tick in the KPPP configuration box and here I am using MD2007 again!
    >
    >Thanks for all your assistance, Old Guy and David.


    Really glad to hear it, and sorry it took this long.

    >Now, I guess I can go through the various files I've edited and see what
    >I can remove to simplify the dial-in process.


    Back in November, I suggested using a dumb script to connect. That would be
    another alternative. You probably want to delete the 'debug' option, as
    that will be putting unneeded stuff into the logs, and kick the port speed
    back up to 115200.

    >Then I can try switching from PAP to CHAP but that might be a little
    >while off.


    I'm not sure how that would be accomplished using KPPP, as one of the
    options your 'dump' report gave was

    Feb 21 17:56:57 www pppd[6314]: -chap^I^I# (from command line)

    which causes pppd to refuse to use CHAP. If you can figure out how to
    stop KPPP from doing that, then the only requirement to use CHAP is for
    the peer to ask for it, AND for /etc/ppp/chap-secrets have the same
    information as /etc/ppp/pap-secrets (username * password). Now your
    logs wasn't showing the peer asking for CHAP

    Feb 21 17:56:57 www pppd[6314]: rcvd [LCP ConfReq id=0x1e 0xa0000>
    ]

    and at this point, the peer wanted to use PAP. Back in November,
    you showed

    Nov 13 20:18:17 localhost pppd[6465]: rcvd [LCP ConfReq id=0x6f
    1524> ]

    and when you ConfNak'ed "auth pap" (and proposed "auth eap" for some
    bizarre reason), the peer did come back with

    Nov 13 20:18:17 localhost pppd[6465]: rcvd [LCP ConfReq id=0x71

    ]

    and there, the peer is asking for CHAP MD5. Is there any particular
    reason you want CHAP instead of PAP? By the way, there is a
    "require-chap" and you do NOT want this option.

    Looking at the dump output again:

    debug^I^I# (from command line)
    -detach^I^I# (from command line)
    dump^I^I# (from /etc/ppp/options)
    noauth^I^I# (from /etc/ppp/options)
    -chap^I^I# (from command line)
    user dxmm@albury.net.au^I^I# (from command line)
    38400^I^I# (from command line)
    lock^I^I# (from /etc/ppp/options)
    mru 1524^I^I# (from command line)
    mtu 1524^I^I# (from command line)
    noipdefault^I^I# (from /etc/ppp/options)
    usepeerdns^I^I# (from command line)
    noccp^I^I# (from /etc/ppp/options)
    nobsdcomp^I^I# (from /etc/ppp/options)
    nodeflate^I^I# (from /etc/ppp/options)

    can probably be reduced to

    -detach^I^I# (from command line)
    user dxmm@albury.net.au^I^I# (from command line)
    115200^I^I# (from command line)
    lock^I^I# (from /etc/ppp/options)
    noipdefault^I^I# (from /etc/ppp/options)
    usepeerdns^I^I# (from command line)
    asyncmap 0xa0000 (from someplace)

    The '(from command line)' is an option set by KPPP, while the
    '(from /etc/ppp/options)' is literally in that file. See the pppd
    man page for an explanation of each. The '-detach' is another bogus
    one from KPPP - the correct option is 'nodetach'.

    Old guy

  6. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Sat, 23 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47bf58c3$0$25987$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> A tick in the KPPP configuration box and here I am using MD2007 again!
    >>
    >> Thanks for all your assistance, Old Guy and David.

    >
    > Really glad to hear it, and sorry it took this long.
    >


    Hey, if I knew what I was doing, it wouldn't have taken this long, but,
    with your help, I got there in the end!

    >> Now, I guess I can go through the various files I've edited and see what
    >> I can remove to simplify the dial-in process.

    >
    > Back in November, I suggested using a dumb script to connect. That would be
    > another alternative. You probably want to delete the 'debug' option, as
    > that will be putting unneeded stuff into the logs, and kick the port speed
    > back up to 115200.
    >
    >> Then I can try switching from PAP to CHAP but that might be a little
    >> while off.

    >
    > I'm not sure how that would be accomplished using KPPP, as one of the
    > options your 'dump' report gave was
    >
    > Feb 21 17:56:57 www pppd[6314]: -chap^I^I# (from command line)
    >
    > which causes pppd to refuse to use CHAP. If you can figure out how to
    > stop KPPP from doing that, then the only requirement to use CHAP is for
    > the peer to ask for it, AND for /etc/ppp/chap-secrets have the same
    > information as /etc/ppp/pap-secrets (username * password). Now your
    > logs wasn't showing the peer asking for CHAP
    >
    > Feb 21 17:56:57 www pppd[6314]: rcvd [LCP ConfReq id=0x1e > 0xa0000>
    > ]
    >
    > and at this point, the peer wanted to use PAP. Back in November,
    > you showed
    >
    > Nov 13 20:18:17 localhost pppd[6465]: rcvd [LCP ConfReq id=0x6f
    > > 1524> ]
    >
    > and when you ConfNak'ed "auth pap" (and proposed "auth eap" for some
    > bizarre reason), the peer did come back with
    >
    > Nov 13 20:18:17 localhost pppd[6465]: rcvd [LCP ConfReq id=0x71
    >
    > ]
    >
    > and there, the peer is asking for CHAP MD5. Is there any particular
    > reason you want CHAP instead of PAP?


    My ISP! He has his server system set up, natively, to work CHAP and has
    had to alter things so I could work PAP cause I couldn't get chap
    working, first time round.

    By the way, there is a
    > "require-chap" and you do NOT want this option.
    >
    > Looking at the dump output again:
    >
    > debug^I^I# (from command line)
    > -detach^I^I# (from command line)
    > dump^I^I# (from /etc/ppp/options)
    > noauth^I^I# (from /etc/ppp/options)
    > -chap^I^I# (from command line)
    > user dxmm@albury.net.au^I^I# (from command line)
    > 38400^I^I# (from command line)
    > lock^I^I# (from /etc/ppp/options)
    > mru 1524^I^I# (from command line)
    > mtu 1524^I^I# (from command line)
    > noipdefault^I^I# (from /etc/ppp/options)
    > usepeerdns^I^I# (from command line)
    > noccp^I^I# (from /etc/ppp/options)
    > nobsdcomp^I^I# (from /etc/ppp/options)
    > nodeflate^I^I# (from /etc/ppp/options)
    >
    > can probably be reduced to
    >
    > -detach^I^I# (from command line)
    > user dxmm@albury.net.au^I^I# (from command line)
    > 115200^I^I# (from command line)
    > lock^I^I# (from /etc/ppp/options)
    > noipdefault^I^I# (from /etc/ppp/options)
    > usepeerdns^I^I# (from command line)
    > asyncmap 0xa0000 (from someplace)
    >
    > The '(from command line)' is an option set by KPPP, while the
    > '(from /etc/ppp/options)' is literally in that file. See the pppd
    > man page for an explanation of each. The '-detach' is another bogus
    > one from KPPP - the correct option is 'nodetach'.
    >
    > Old guy


    I'll take it slowly, and see how I go.

    Tks

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  7. Re: Using PPP to dial-in to ISP

    On Sat, 23 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47bf7052$0$26083$88260bb3@free.teranews.com>, Daniel wrote:

    >Moe Trin wrote:


    >> Really glad to hear it, and sorry it took this long.

    >
    >Hey, if I knew what I was doing, it wouldn't have taken this long, but,
    >with your help, I got there in the end!


    Actually, the asyncmap problem is known to those who also followed
    ppp questions in comp.os.linux.networking and comp.protocols.ppp,
    but isn't well known otherwise. The 'receive-all' option was
    another way around these b0rken peers, but was only introduced in
    ppp-2.3.7 back in 1999.

    >> Is there any particular reason you want CHAP instead of PAP?

    >
    >My ISP! He has his server system set up, natively, to work CHAP and has
    >had to alter things so I could work PAP cause I couldn't get chap
    >working, first time round.


    Oh, OK - again, I'm not using KPPP, but there should be a tic-box as
    to which authentication scheme you are using. As mentioned, pppd
    doesn't care by default, and if the peer proposes method $FOO, it
    will use that method (PAP, CHAP-MD5, CHAP-MSv1, CHAP-MSv2, EAP) if
    the appropriate secrets file (/etc/ppp/{pap|chap|sap}-secrets) has
    an appropriate entry. KPPP seems to be set to restrict this, but
    I'm not the expert on KPPP.

    >I'll take it slowly, and see how I go.


    The options I listed should be all that is needed. For giggles,

    [compton ~]$ cat /usr/local/bin/dialin
    #!/bin/bash
    exec /usr/sbin/pppd connect "/usr/sbin/chat -f /etc/ppp/dialscript" \
    lock defaultroute noipdefault nodetach /dev/modem 115200 \
    asyncmap 0xa0000 user dxmm@albury.net.au
    [compton ~]$

    There must not be anything after the \ in those lines. This is also
    assuming that /dev/modem is a link that points to where your modem is
    really hiding. Note that some of these options may be included in
    /etc/ppp/options, and if so, don't need to be repeated here.

    [compton ~]$ cat /etc/ppp/dialscript
    ABORT BUSY ABORT 'NO CARRIER' "" AT&F0 OK ATDT2662902 CONNECT \d\c
    [compton ~]$

    with the corrected phone number, and the right stuff in the secrets
    file[s], and /etc/resolv.conf, that's about all it should take.

    Old guy

  8. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Sat, 23 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47bf7052$0$26083$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> Moe Trin wrote:

    >
    >>> Really glad to hear it, and sorry it took this long.

    >> Hey, if I knew what I was doing, it wouldn't have taken this long, but,
    >> with your help, I got there in the end!

    >
    > Actually, the asyncmap problem is known to those who also followed
    > ppp questions in comp.os.linux.networking and comp.protocols.ppp,
    > but isn't well known otherwise. The 'receive-all' option was
    > another way around these b0rken peers, but was only introduced in
    > ppp-2.3.7 back in 1999.
    >
    >>> Is there any particular reason you want CHAP instead of PAP?

    >> My ISP! He has his server system set up, natively, to work CHAP and has
    >> had to alter things so I could work PAP cause I couldn't get chap
    >> working, first time round.

    >


    More correctly, I believe he stores all the passwords for his subscriber
    accounts after they have been MD5'ed, so he doesn't actually have the
    passwords anywhere that they can be accessed. So, then when his
    subscriber logs in, he sends out the MD5 "word" for the subscriber to
    mash with the subscribers own copy of the password, and the result is
    feed back and comparer ... the password is never "out there" in plain.

    > Oh, OK - again, I'm not using KPPP, but there should be a tic-box as
    > to which authentication scheme you are using. As mentioned, pppd
    > doesn't care by default, and if the peer proposes method $FOO, it
    > will use that method (PAP, CHAP-MD5, CHAP-MSv1, CHAP-MSv2, EAP) if
    > the appropriate secrets file (/etc/ppp/{pap|chap|sap}-secrets) has
    > an appropriate entry. KPPP seems to be set to restrict this, but
    > I'm not the expert on KPPP.
    >
    >> I'll take it slowly, and see how I go.

    >
    > The options I listed should be all that is needed. For giggles,
    >
    > [compton ~]$ cat /usr/local/bin/dialin
    > #!/bin/bash
    > exec /usr/sbin/pppd connect "/usr/sbin/chat -f /etc/ppp/dialscript" \
    > lock defaultroute noipdefault nodetach /dev/modem 115200 \
    > asyncmap 0xa0000 user dxmm@albury.net.au
    > [compton ~]$
    >


    Don't have a /usr/local/bin/dialin but do have a /usr/local/bin/!dialin,
    note the ! before dialin. This contains:-
    #!/bin/bash
    exec /usr/sbin/pppd connect "/usr/sbin/chat -f /etc/ppp/dialscript" lock \
    defaultroute noipdefault nodetach /dev/modem 115200 crtscts nomrru\
    asyncmap 0xa0000 user dxmm@albury.net.au

    > There must not be anything after the \ in those lines. This is also
    > assuming that /dev/modem is a link that points to where your modem is
    > really hiding. Note that some of these options may be included in
    > /etc/ppp/options, and if so, don't need to be repeated here.
    >
    > [compton ~]$ cat /etc/ppp/dialscript
    > ABORT BUSY ABORT 'NO CARRIER' "" AT&F0 OK ATDT2662902 CONNECT \d\c
    > [compton ~]$
    >


    [daniel@P013 ~]$ cat /etc/ppp/dialscript
    ABORT BUSY ABORT 'NO CARRIER' "" AT&F0 OK ATDT0198331352 CONNECT \d\c

    > with the corrected phone number, and the right stuff in the secrets
    > file[s], and /etc/resolv.conf, that's about all it should take.
    >
    > Old guy


    O.K. Two more questions, if I could:-

    1. Why am I now getting [daniel@P013 ~]$ rather than the more normal
    [daniel@local]$, and,

    2. In my SeaMonkey browser, I can go to google.com, I can go to
    http://www.gowildballooning.com.au/baln_flight.html, I can even go to
    http://weather.albury.net.au/ (page banner is missing, though), but I
    cannot get any of my ISP's pages, i.e. www.albury.net.au. Have I done
    something to disable that domain in Mandriva?? (I've even turned off the
    Firewall, still no go, but works for Win98 SeaMonkey!)

    Tks

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  9. Re: Using PPP to dial-in to ISP

    On Mon, 25 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47c2013d$0$26114$88260bb3@free.teranews.com>, Daniel wrote:

    >More correctly, I believe he stores all the passwords for his subscriber
    >accounts after they have been MD5'ed, so he doesn't actually have the
    >passwords anywhere that they can be accessed. So, then when his
    >subscriber logs in, he sends out the MD5 "word" for the subscriber to
    >mash with the subscribers own copy of the password, and the result is
    >feed back and comparer ... the password is never "out there" in plain.


    1334 PPP Authentication Protocols. B. Lloyd, W. Simpson. October 1992.
    (Format: TXT=33248 bytes) (Obsoleted by RFC1994) (Status: PROPOSED
    STANDARD)

    1994 PPP Challenge Handshake Authentication Protocol (CHAP). W.
    Simpson. August 1996. (Format: TXT=24094 bytes) (Obsoletes RFC1334)
    (Updated by RFC2484) (Status: DRAFT STANDARD)

    There are also two microsoft (incompatible) "standards". While CHAP has
    obsoleted the earlier PAP, the later remains far more common than CHAP
    despite the improved security it offers.

    >Don't have a /usr/local/bin/dialin but do have a /usr/local/bin/!dialin,
    >note the ! before dialin


    Hmm, how did you get that one? You didn't seem to have the typo back in
    November/December. ;-)

    >This contains:-
    >#!/bin/bash
    >exec /usr/sbin/pppd connect "/usr/sbin/chat -f /etc/ppp/dialscript" lock \
    >defaultroute noipdefault nodetach /dev/modem 115200 crtscts nomrru\


    Where did I get the 'nomrru' option... That should be 'nomrp' or
    'nomultilink' - but only if the peer is having problems with it.

    >asyncmap 0xa0000 user dxmm@albury.net.au


    Other than the nomrru - that's fine.

    >[daniel@P013 ~]$ cat /etc/ppp/dialscript
    >ABORT BUSY ABORT 'NO CARRIER' "" AT&F0 OK ATDT0198331352 CONNECT \d\c


    Good

    >O.K. Two more questions, if I could:-
    >
    >1. Why am I now getting [daniel@P013 ~]$ rather than the more normal
    >[daniel@local]$, and,


    Oh, I just love these helper programs that know what you really want
    to do. This is a service of KPPP which is renaming your computer when
    you connect. Another d4mn tic-box somewhere in the KPPP setup. A problem
    this causes is that by renaming your system while X is running, you
    prevent any new services from starting, because you don't have
    permission to use the desktop on this "other" host. Normally, your
    hostname is set in the file /etc/sysconfig/network with the "HOSTNAME="
    line. You should also have "NETWORKING_IPV6=no", "NOZEROCONF=yes" and
    "NEEDHOSTNAME=no" in that same file.

    >2. In my SeaMonkey browser, I can go to google.com, I can go to
    >http://www.gowildballooning.com.au/baln_flight.html, I can even go to
    >http://weather.albury.net.au/ (page banner is missing, though), but I
    >cannot get any of my ISP's pages, i.e. www.albury.net.au.


    I don't use SeaMonkey, so I can't say what that might be, but it vaguely
    sounds like some kind of browser configuration. The www.albury.net.au
    page comes up here with no problems, although it seems to be grabbing
    stuff from all over the place - multiple links within the page. Ah
    the missing banner seems to be from http://www.albury.net.au/images/
    and I wonder why you can't get that. Does the hostname resolve? Here
    in the US at 2130UTC/25 Feb, I'm told it is a nickname for
    ali-syd-1.albury.net.au, and is at 202.3.36.15. APNIC tells me that
    address actually belongs to worley.org.au in Campbelltown NSW.

    >Have I done something to disable that domain in Mandriva?? (I've even
    >turned off the Firewall, still no go, but works for Win98 SeaMonkey!)


    I can't see an obvious reason. Again, looking from the US, the
    weather.albury.net.au is a nickname for terran.albury.net.au and is at
    203.15.244.18, which APNIC tells me belongs to Albury Local Internet Pty
    Ltd in Lavington, NSW. Are seeing any error messages? Have you tried
    another browser, such as lynx or links? Another wild shot would be
    a 'search' or 'domain' directive in /etc/resolv.conf.

    Old guy

  10. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Mon, 25 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47c2013d$0$26114$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> More correctly, I believe he stores all the passwords for his subscriber
    >> accounts after they have been MD5'ed, so he doesn't actually have the
    >> passwords anywhere that they can be accessed. So, then when his
    >> subscriber logs in, he sends out the MD5 "word" for the subscriber to
    >> mash with the subscribers own copy of the password, and the result is
    >> feed back and comparer ... the password is never "out there" in plain.

    >
    > 1334 PPP Authentication Protocols. B. Lloyd, W. Simpson. October 1992.
    > (Format: TXT=33248 bytes) (Obsoleted by RFC1994) (Status: PROPOSED
    > STANDARD)
    >
    > 1994 PPP Challenge Handshake Authentication Protocol (CHAP). W.
    > Simpson. August 1996. (Format: TXT=24094 bytes) (Obsoletes RFC1334)
    > (Updated by RFC2484) (Status: DRAFT STANDARD)
    >
    > There are also two microsoft (incompatible) "standards". While CHAP has
    > obsoleted the earlier PAP, the later remains far more common than CHAP
    > despite the improved security it offers.
    >
    >> Don't have a /usr/local/bin/dialin but do have a /usr/local/bin/!dialin,
    >> note the ! before dialin

    >
    > Hmm, how did you get that one? You didn't seem to have the typo back in
    > November/December. ;-)
    >
    >> This contains:-
    >> #!/bin/bash
    >> exec /usr/sbin/pppd connect "/usr/sbin/chat -f /etc/ppp/dialscript" lock \
    >> defaultroute noipdefault nodetach /dev/modem 115200 crtscts nomrru\

    >
    > Where did I get the 'nomrru' option... That should be 'nomrp' or
    > 'nomultilink' - but only if the peer is having problems with it.
    >
    >> asyncmap 0xa0000 user dxmm@albury.net.au

    >
    > Other than the nomrru - that's fine.
    >


    O.K., do I just rename this and make the edits you mention here??

    >> [daniel@P013 ~]$ cat /etc/ppp/dialscript
    >> ABORT BUSY ABORT 'NO CARRIER' "" AT&F0 OK ATDT0198331352 CONNECT \d\c

    >
    > Good
    >
    >> O.K. Two more questions, if I could:-
    >>
    >> 1. Why am I now getting [daniel@P013 ~]$ rather than the more normal
    >> [daniel@local]$, and,

    >
    > Oh, I just love these helper programs that know what you really want
    > to do. This is a service of KPPP which is renaming your computer when
    > you connect. Another d4mn tic-box somewhere in the KPPP setup. A problem
    > this causes is that by renaming your system while X is running, you
    > prevent any new services from starting, because you don't have
    > permission to use the desktop on this "other" host. Normally, your
    > hostname is set in the file /etc/sysconfig/network with the "HOSTNAME="
    > line. You should also have "NETWORKING_IPV6=no", "NOZEROCONF=yes" and
    > "NEEDHOSTNAME=no" in that same file.
    >


    /etc/sysconfig/network just has:-
    NETWORKING=yes

    >> 2. In my SeaMonkey browser, I can go to google.com, I can go to
    >> http://www.gowildballooning.com.au/baln_flight.html, I can even go to
    >> http://weather.albury.net.au/ (page banner is missing, though), but I
    >> cannot get any of my ISP's pages, i.e. www.albury.net.au.

    >
    > I don't use SeaMonkey, so I can't say what that might be, but it vaguely
    > sounds like some kind of browser configuration. The www.albury.net.au
    > page comes up here with no problems, although it seems to be grabbing
    > stuff from all over the place - multiple links within the page. Ah
    > the missing banner seems to be from http://www.albury.net.au/images/
    > and I wonder why you can't get that. Does the hostname resolve? Here
    > in the US at 2130UTC/25 Feb, I'm told it is a nickname for
    > ali-syd-1.albury.net.au, and is at 202.3.36.15. APNIC tells me that
    > address actually belongs to worley.org.au in Campbelltown NSW.
    >


    [daniel@P013 ~]$ host albury.net.au
    albury.net.au has address 202.3.36.15
    albury.net.au mail is handled by 10 mail3.albury.net.au.
    albury.net.au mail is handled by 20 mail.albury.net.au.
    [daniel@P013 ~]$

    Did I mention that these problems started when my ISP had to change from
    a local call number to a National Call number routed through Sydney -
    Campbelltown is, more or less, a suburb of Sydney.

    >> Have I done something to disable that domain in Mandriva?? (I've even
    >> turned off the Firewall, still no go, but works for Win98 SeaMonkey!)

    >
    > I can't see an obvious reason. Again, looking from the US, the
    > weather.albury.net.au is a nickname for terran.albury.net.au and is at
    > 203.15.244.18, which APNIC tells me belongs to Albury Local Internet Pty
    > Ltd in Lavington, NSW. Are seeing any error messages? Have you tried
    > another browser, such as lynx or links? Another wild shot would be
    > a 'search' or 'domain' directive in /etc/resolv.conf.
    >
    > Old guy


    When I use SeaMonkey to try to get to www.albury.net.au, it says "Fail
    to Connect -- The connection was refused when attempting to contact
    www.albury.net.au."

    Just tried Konqueror and it's not getting out at all, will have to check
    the set-up. Mozilla Firefox ver 1.5.0.7 can get to Google, the weather
    but not to www.albury.net.au It's Unable to connect.

    --
    Posted via a free Usenet account from http://www.teranews.com


  11. Re: Using PPP to dial-in to ISP

    On Tue, 26 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47c3de81$0$18851$88260bb3@free.teranews.com>, Daniel wrote:

    >Moe Trin wrote:


    >> Other than the nomrru - that's fine.

    >
    >O.K., do I just rename this and make the edits you mention here??


    That would be fine

    >> Normally, your hostname is set in the file /etc/sysconfig/network
    >> with the "HOSTNAME=" line. You should also have "NETWORKING_IPV6=no",
    >> "NOZEROCONF=yes" and "NEEDHOSTNAME=no" in that same file.

    >
    >/etc/sysconfig/network just has:-
    >NETWORKING=yes


    Hmmm. Bit Twister has recommended the following for a system that has
    a NIC and accessing the Internet through a device there:

    --------------
    $ cat /etc/sysconfig/network
    NETWORKING_IPV6=no <---- speedup DNS lookups
    NOZEROCONF=yes <---- no doze lookups needed
    GATEWAYDEV=eth0
    GATEWAY=192.168.2.1
    HOSTNAME=wb.home.invalid <---- sets node name on boot (FQDN)
    NEEDHOSTNAME=no <---- keeps DHCP value from overwriting it
    NETWORKING=yes
    ---------------

    but you say you have removed the NIC as you have no Ethernet. I suspect
    you want to have this file set to

    --------------
    NETWORKING_IPV6=no
    NOZEROCONF=yes
    HOSTNAME=whatever.you.call.the.system
    NEEDHOSTNAME=no
    NETWORKING=yes
    --------------

    >[daniel@P013 ~]$ host albury.net.au
    >albury.net.au has address 202.3.36.15
    >albury.net.au mail is handled by 10 mail3.albury.net.au.
    >albury.net.au mail is handled by 20 mail.albury.net.au.
    >[daniel@P013 ~]$


    Does www.albury.net.au resolve as well?

    >Did I mention that these problems started when my ISP had to change from
    >a local call number to a National Call number routed through Sydney -
    >Campbelltown is, more or less, a suburb of Sydney.


    I don't see what that would have to do with it - you say that windoze
    works, and I'm assuming you are using the same phone number, which
    sort of implies the same general networking values. Looking at the
    windoze side... OK win98:

    C:\WINDOWS\HOSTS (not hosts.sam) is windoze9x/me hosts file.
    If you are using Windows 9X, ME then winipcfg and the more button will
    tell you what specific ISP stuff you are using, although they do try
    to confuse you with techno-babble.

    On the Linux side, look at /etc/hosts (if all you have is a loopback
    interface, only 127.0.0.1 will be needed), /etc/resolv.conf (should
    have both nameservers listed - no search or no domain lines), and
    the output of /sbin/ifconfig and /sbin/route - which as I recall now
    looks normal.

    >When I use SeaMonkey to try to get to www.albury.net.au, it says "Fail
    >to Connect -- The connection was refused when attempting to contact
    >www.albury.net.au."
    >
    >Just tried Konqueror and it's not getting out at all, will have to check
    >the set-up. Mozilla Firefox ver 1.5.0.7 can get to Google, the weather
    >but not to www.albury.net.au It's Unable to connect.


    Hmmm... I'm a network guy, and my next trick would be to run a
    packet sniffer to see what you are trying to connect to. Perhaps
    something as simple as

    /usr/sbin/tcpdump -n -i ppp0

    but be advised that this can produce a huge amount of data. What you
    are looking for is your system sending a UDP datagram to a nameserver
    to identify the IP of the hostname you want to connect to, a response
    telling you this IP, and then a connection attempt to that host. Does
    the host respond with something other than a packet with the letter
    'R' in the header, and a 'win 0'? An example:

    09:10:48.80 192.168.10.54.9562 > 192.168.10.31.80: S
    1130717347:1130717347(0) win 512
    09:10:48.96 192.168.10.31.80 > 192.168.10.54.9562: R 0:0(0) ack
    1130717348 win 0

    here, 192.168.10.54 tried to connect to the web server on 192.168.10.31
    (the S at the end of the first line means the start of a connection).
    That host isn't running a web server, and replied 'R" (RESET - meaning
    "go away kid, you're bothering me") and the 'win 0' which means this
    conversation is finished - don't talk to me.

    A successful connection is going to have a LOT of exchanges of packets.
    Just starting a connection takes three packets and that's before it
    starts asking about a URL. When I looked at www.albury.net.au, I saw
    it getting at least ten images (each being 6 to 10 packets), alone.
    There were 46 URLs on the primary page, some of which autoload. Can
    you say "lots of packets"? ;-)

    What you are looking for is connection attempts (like the first one
    at 09:10:48.80 above) that do NOT get a response - they will
    probably be repeated three times a few seconds apart.

    Old guy

  12. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Tue, 26 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47c3de81$0$18851$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> Moe Trin wrote:

    >
    >>> Other than the nomrru - that's fine.

    >> O.K., do I just rename this and make the edits you mention here??

    >
    > That would be fine
    >
    >>> Normally, your hostname is set in the file /etc/sysconfig/network
    >>> with the "HOSTNAME=" line. You should also have "NETWORKING_IPV6=no",
    >>> "NOZEROCONF=yes" and "NEEDHOSTNAME=no" in that same file.

    >> /etc/sysconfig/network just has:-
    >> NETWORKING=yes

    >
    > Hmmm. Bit Twister has recommended the following for a system that has
    > a NIC and accessing the Internet through a device there:
    >
    > --------------
    > $ cat /etc/sysconfig/network
    > NETWORKING_IPV6=no <---- speedup DNS lookups
    > NOZEROCONF=yes <---- no doze lookups needed
    > GATEWAYDEV=eth0
    > GATEWAY=192.168.2.1
    > HOSTNAME=wb.home.invalid <---- sets node name on boot (FQDN)
    > NEEDHOSTNAME=no <---- keeps DHCP value from overwriting it
    > NETWORKING=yes
    > ---------------
    >
    > but you say you have removed the NIC as you have no Ethernet. I suspect
    > you want to have this file set to
    >
    > --------------
    > NETWORKING_IPV6=no
    > NOZEROCONF=yes
    > HOSTNAME=whatever.you.call.the.system
    > NEEDHOSTNAME=no
    > NETWORKING=yes
    > --------------
    >
    >> [daniel@P013 ~]$ host albury.net.au
    >> albury.net.au has address 202.3.36.15
    >> albury.net.au mail is handled by 10 mail3.albury.net.au.
    >> albury.net.au mail is handled by 20 mail.albury.net.au.
    >> [daniel@P013 ~]$

    >
    > Does www.albury.net.au resolve as well?
    >
    >> Did I mention that these problems started when my ISP had to change from
    >> a local call number to a National Call number routed through Sydney -
    >> Campbelltown is, more or less, a suburb of Sydney.

    >
    > I don't see what that would have to do with it - you say that windoze
    > works, and I'm assuming you are using the same phone number, which
    > sort of implies the same general networking values. Looking at the
    > windoze side... OK win98:
    >
    > C:\WINDOWS\HOSTS (not hosts.sam) is windoze9x/me hosts file.
    > If you are using Windows 9X, ME then winipcfg and the more button will
    > tell you what specific ISP stuff you are using, although they do try
    > to confuse you with techno-babble.
    >
    > On the Linux side, look at /etc/hosts (if all you have is a loopback
    > interface, only 127.0.0.1 will be needed), /etc/resolv.conf (should
    > have both nameservers listed - no search or no domain lines), and
    > the output of /sbin/ifconfig and /sbin/route - which as I recall now
    > looks normal.
    >
    >> When I use SeaMonkey to try to get to www.albury.net.au, it says "Fail
    >> to Connect -- The connection was refused when attempting to contact
    >> www.albury.net.au."
    >>
    >> Just tried Konqueror and it's not getting out at all, will have to check
    >> the set-up. Mozilla Firefox ver 1.5.0.7 can get to Google, the weather
    >> but not to www.albury.net.au It's Unable to connect.

    >
    > Hmmm... I'm a network guy, and my next trick would be to run a
    > packet sniffer to see what you are trying to connect to. Perhaps
    > something as simple as
    >
    > /usr/sbin/tcpdump -n -i ppp0
    >
    > but be advised that this can produce a huge amount of data. What you
    > are looking for is your system sending a UDP datagram to a nameserver
    > to identify the IP of the hostname you want to connect to, a response
    > telling you this IP, and then a connection attempt to that host. Does
    > the host respond with something other than a packet with the letter
    > 'R' in the header, and a 'win 0'? An example:
    >
    > 09:10:48.80 192.168.10.54.9562 > 192.168.10.31.80: S
    > 1130717347:1130717347(0) win 512
    > 09:10:48.96 192.168.10.31.80 > 192.168.10.54.9562: R 0:0(0) ack
    > 1130717348 win 0
    >
    > here, 192.168.10.54 tried to connect to the web server on 192.168.10.31
    > (the S at the end of the first line means the start of a connection).
    > That host isn't running a web server, and replied 'R" (RESET - meaning
    > "go away kid, you're bothering me") and the 'win 0' which means this
    > conversation is finished - don't talk to me.
    >
    > A successful connection is going to have a LOT of exchanges of packets.
    > Just starting a connection takes three packets and that's before it
    > starts asking about a URL. When I looked at www.albury.net.au, I saw
    > it getting at least ten images (each being 6 to 10 packets), alone.
    > There were 46 URLs on the primary page, some of which autoload. Can
    > you say "lots of packets"? ;-)
    >
    > What you are looking for is connection attempts (like the first one
    > at 09:10:48.80 above) that do NOT get a response - they will
    > probably be repeated three times a few seconds apart.
    >
    > Old guy


    Sorry for the delay, Old Guy, but kppp has been closing down (error code
    16) before I had a chance to reply to you, just to let you know that I
    had read this.

    With my replying, I'll now have a copy of your response for me to look
    at and work through in both Win98 and MD2007.

    Will respond in a couple of days.

    Tks.

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  13. Re: Using PPP to dial-in to ISP

    On Sun, 02 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47ca66ad$0$16010$88260bb3@free.teranews.com>, Daniel wrote:

    >Sorry for the delay, Old Guy, but kppp has been closing down (error
    >code 16) before I had a chance to reply to you, just to let you know
    >that I had read this.


    Turn your debugging back on - and recheck the logs. Near the end of
    the pppd man page, there is a section "EXIT STATUS" with the rather
    simple-minded explanations of those codes. In this case:

    16 The link was terminated by the modem hanging up.

    That usually means that something related to a wrong option was used,
    sometimes the wrong authentication mechanism, sometimes a failure of
    the 'connect' script (dialing the modem). The peer didn't like what
    you said, and hung up the phone.

    Old guy

  14. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Sun, 02 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47ca66ad$0$16010$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> Sorry for the delay, Old Guy, but kppp has been closing down (error
    >> code 16) before I had a chance to reply to you, just to let you know
    >> that I had read this.

    >
    > Turn your debugging back on - and recheck the logs. Near the end of
    > the pppd man page, there is a section "EXIT STATUS" with the rather
    > simple-minded explanations of those codes. In this case:
    >
    > 16 The link was terminated by the modem hanging up.
    >
    > That usually means that something related to a wrong option was used,
    > sometimes the wrong authentication mechanism, sometimes a failure of
    > the 'connect' script (dialing the modem). The peer didn't like what
    > you said, and hung up the phone.
    >
    > Old guy


    Yes, Moe, but this is not at my behest! I've suspected the line level
    may, at times, be marginal, and sometimes the line level falls below the
    modems minimum usable level, so the modem closes the connection.

    Is this far off the mark??

    Daniel

    --
    Posted via a free Usenet account from http://www.teranews.com


  15. Re: Using PPP to dial-in to ISP

    Daniel wrote:
    > Moe Trin wrote:
    >> On Tue, 26 Feb 2008, in the Usenet newsgroup alt.os.linux.mandriva, in
    >> article
    >> <47c3de81$0$18851$88260bb3@free.teranews.com>, Daniel wrote:
    >>
    >>> Moe Trin wrote:

    >>
    >>>> Other than the nomrru - that's fine.
    >>> O.K., do I just rename this and make the edits you mention here??

    >>
    >> That would be fine
    >>
    >>>> Normally, your hostname is set in the file /etc/sysconfig/network
    >>>> with the "HOSTNAME=" line. You should also have "NETWORKING_IPV6=no",
    >>>> "NOZEROCONF=yes" and "NEEDHOSTNAME=no" in that same file.
    >>> /etc/sysconfig/network just has:-
    >>> NETWORKING=yes

    >>
    >> Hmmm. Bit Twister has recommended the following for a system that has
    >> a NIC and accessing the Internet through a device there:
    >>
    >> --------------
    >> $ cat /etc/sysconfig/network
    >> NETWORKING_IPV6=no <---- speedup DNS lookups
    >> NOZEROCONF=yes <---- no doze lookups needed
    >> GATEWAYDEV=eth0
    >> GATEWAY=192.168.2.1
    >> HOSTNAME=wb.home.invalid <---- sets node name on boot (FQDN)
    >> NEEDHOSTNAME=no <---- keeps DHCP value from overwriting it
    >> NETWORKING=yes
    >> ---------------
    >>
    >> but you say you have removed the NIC as you have no Ethernet.


    Yeap!

    I suspect
    >> you want to have this file set to
    >>
    >> --------------
    >> NETWORKING_IPV6=no
    >> NOZEROCONF=yes
    >> HOSTNAME=whatever.you.call.the.system
    >> NEEDHOSTNAME=no
    >> NETWORKING=yes
    >> --------------
    >>


    for "HOSTNAME", where you have "whatever.you.call.the.system", you mean
    my system profile??? Or something else???

    >>> [daniel@P013 ~]$ host albury.net.au
    >>> albury.net.au has address 202.3.36.15
    >>> albury.net.au mail is handled by 10 mail3.albury.net.au.
    >>> albury.net.au mail is handled by 20 mail.albury.net.au.
    >>> [daniel@P013 ~]$

    >>
    >> Does www.albury.net.au resolve as well?
    >>
    >>> Did I mention that these problems started when my ISP had to change from
    >>> a local call number to a National Call number routed through Sydney -
    >>> Campbelltown is, more or less, a suburb of Sydney.

    >>
    >> I don't see what that would have to do with it - you say that windoze
    >> works, and I'm assuming you are using the same phone number, which
    >> sort of implies the same general networking values. Looking at the
    >> windoze side... OK win98:
    >>
    >> C:\WINDOWS\HOSTS (not hosts.sam) is windoze9x/me hosts file.
    >> If you are using Windows 9X, ME then winipcfg and the more button will
    >> tell you what specific ISP stuff you are using, although they do try
    >> to confuse you with techno-babble.
    >>


    On Windows 98.......
    No hosts file anywhere!

    If I run winipcfg, the "More" button shows host name of "DXMM" and Node
    type "Broadcast". Interestingly enough in the Ethernet Adapter
    Information the only Adapter I have listed is "PPP Adapter" with the
    adapter address of 44-45-53-54-00-00 and IP address of 0.0.0.0 and
    subnet Mask of 0.0.0.0 and the DHCP Server is 255.255.255.255

    >> On the Linux side, look at /etc/hosts (if all you have is a loopback
    >> interface, only 127.0.0.1 will be needed),


    /etc/hosts
    127.0.0.1 www.albury.net.au www localhost dxmm@albury.net.au
    dxmm@albury dxmm

    I'm betting this is this why I cannot get to my ISP's home page -
    www.albury.net.au

    /etc/resolv.conf

    search net.au
    # search domain
    nameserver 202.3.36.10
    nameserver 202.3.36.11
    # nameserver 127.0.0.1
    # nameserver 127.0.0.1
    # nameserver 127.0.0.1
    # $Id: resolv.conf,v 1.1 2006/06/02 23:25:57 othauvin Exp $
    # Basic resolv.conf


    (should
    >> have both nameservers listed - no search or no domain lines), and
    >> the output of /sbin/ifconfig


    /sbin/ifconfig output whilst not connected/not dialed in:-
    [root@localhost daniel]# /sbin/ifconfig
    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:81 errors:0 dropped:0 overruns:0 frame:0
    TX packets:81 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:5645 (5.5 KiB) TX bytes:5645 (5.5 KiB)

    [root@localhost daniel]#

    and /sbin/route - which as I recall now
    >> looks normal.
    >>

    /sbin/route, again not dialed in:-

    [root@localhost daniel]# /sbin/route
    Kernel IP routing table
    Destination Gateway Genmask Flags Metric Ref Use
    Iface
    169.254.0.0 * 255.255.0.0 U 0 0 0 lo
    127.0.0.0 * 255.0.0.0 U 0 0 0 lo
    [root@localhost daniel]#

    >>> When I use SeaMonkey to try to get to www.albury.net.au, it says "Fail
    >>> to Connect -- The connection was refused when attempting to contact
    >>> www.albury.net.au."
    >>>


    removed www.albury.net.au from hosts file ... see how that goes!

    >>> Just tried Konqueror and it's not getting out at all, will have to check
    >>> the set-up. Mozilla Firefox ver 1.5.0.7 can get to Google, the weather
    >>> but not to www.albury.net.au It's Unable to connect.

    >>



    I haven't done this stuff that follows......little bits...see how I go!!

    >> Hmmm... I'm a network guy, and my next trick would be to run a
    >> packet sniffer to see what you are trying to connect to. Perhaps
    >> something as simple as
    >>
    >> /usr/sbin/tcpdump -n -i ppp0
    >>
    >> but be advised that this can produce a huge amount of data. What you
    >> are looking for is your system sending a UDP datagram to a nameserver
    >> to identify the IP of the hostname you want to connect to, a response
    >> telling you this IP, and then a connection attempt to that host. Does
    >> the host respond with something other than a packet with the letter
    >> 'R' in the header, and a 'win 0'? An example:
    >>
    >> 09:10:48.80 192.168.10.54.9562 > 192.168.10.31.80: S
    >> 1130717347:1130717347(0) win 512
    >> 09:10:48.96 192.168.10.31.80 > 192.168.10.54.9562: R 0:0(0) ack
    >> 1130717348 win 0
    >>
    >> here, 192.168.10.54 tried to connect to the web server on 192.168.10.31
    >> (the S at the end of the first line means the start of a connection).
    >> That host isn't running a web server, and replied 'R" (RESET - meaning
    >> "go away kid, you're bothering me") and the 'win 0' which means this
    >> conversation is finished - don't talk to me.
    >>
    >> A successful connection is going to have a LOT of exchanges of packets.
    >> Just starting a connection takes three packets and that's before it
    >> starts asking about a URL. When I looked at www.albury.net.au, I saw
    >> it getting at least ten images (each being 6 to 10 packets), alone.
    >> There were 46 URLs on the primary page, some of which autoload. Can
    >> you say "lots of packets"? ;-)
    >>
    >> What you are looking for is connection attempts (like the first one
    >> at 09:10:48.80 above) that do NOT get a response - they will
    >> probably be repeated three times a few seconds apart.
    >>
    >> Old guy

    >
    > Sorry for the delay, Old Guy, but kppp has been closing down (error code
    > 16) before I had a chance to reply to you, just to let you know that I
    > had read this.
    >
    > With my replying, I'll now have a copy of your response for me to look
    > at and work through in both Win98 and MD2007.
    >
    > Will respond in a couple of days.
    >
    > Tks.
    >
    > Daniel
    >




    --
    Posted via a free Usenet account from http://www.teranews.com


  16. Re: Using PPP to dial-in to ISP

    On Mon, 03 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47cbddb9$0$26047$88260bb3@free.teranews.com>, Daniel wrote:

    >Moe Trin wrote:


    >> Turn your debugging back on - and recheck the logs. Near the end of
    >> the pppd man page, there is a section "EXIT STATUS" with the rather
    >> simple-minded explanations of those codes.


    >> The peer didn't like what you said, and hung up the phone.


    >Yes, Moe, but this is not at my behest! I've suspected the line level
    >may, at times, be marginal, and sometimes the line level falls below
    >the modems minimum usable level, so the modem closes the connection.


    That _should_ only occur after the connection has been established,
    as otherwise pppd would report a different exit code (probably 8).
    Lessee, back in November, you reported this as a "Motorola Lifestyle"
    modem, and the only such modem I'm aware of is an ancient 28.8k ISA
    Internal modem that used a Motorola chip set of some kind. Do you
    have a manual for it? Most modems have a post-call diagnostics
    command of some kind - unfortunately the command and data format is
    not standardized across models from the same manufacturer, much less
    from different ones. One of the things these commands can report is
    the signal strength, and the signal to noise ratio. Another thing often
    reported it the reason the link was terminated.

    >Is this far off the mark??


    It's certainly not impossible, but most modems will "work" with a pretty
    bad signal level - something like 1/100 the level of a "normal" voice
    connection. This would be VERY noticeable. Can you hear the modem
    negotiations (the sounds the two modems make to each other when they
    are trying to establish a connection)?

    Old guy

  17. Re: Using PPP to dial-in to ISP

    On Mon, 03 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47cbe1c0$0$26117$88260bb3@free.teranews.com>, Daniel wrote:

    >Daniel wrote:


    >> Moe Trin wrote:


    >>> I suspect you want to have this file set to
    >>> --------------
    >>> NETWORKING_IPV6=no
    >>> NOZEROCONF=yes
    >>> HOSTNAME=whatever.you.call.the.system
    >>> NEEDHOSTNAME=no
    >>> NETWORKING=yes
    >>> --------------


    >for "HOSTNAME", where you have "whatever.you.call.the.system", you mean
    >my system profile??? Or something else???


    This is meant to be the name that you want to call the system. For
    example, this system is called 'compton.phx.az.us' as shown by the
    prompt:

    [compton ~]$ grep HOSTNAME /etc/sysconfig/network
    HOSTNAME=compton.phx.az.us
    [compton ~]$

    >>> C:\WINDOWS\HOSTS (not hosts.sam) is windoze9x/me hosts file.
    >>> If you are using Windows 9X, ME then winipcfg and the more button will
    >>> tell you what specific ISP stuff you are using, although they do try
    >>> to confuse you with techno-babble.


    >On Windows 98.......
    >No hosts file anywhere!


    I got rid of windoze back in 1992 before they invented networking, so
    I'm going on what others have posted.

    >If I run winipcfg, the "More" button shows host name of "DXMM" and Node
    >type "Broadcast".


    Technically, "DXMM" would be the short name, as a hostname should have
    one or more 'dots' in it - such as dxmm.albury.net.au.invalid (the
    'invalid' suffix should be included if that's not a real domain name
    and there is even a tiny chance that there could be confusion).

    >Interestingly enough in the Ethernet Adapter Information the only
    >Adapter I have listed is "PPP Adapter" with the adapter address of
    >44-45-53-54-00-00 and IP address of 0.0.0.0 and subnet Mask of 0.0.0.0
    >and the DHCP Server is 255.255.255.255


    Meaningless techno-babble.

    >>> On the Linux side, look at /etc/hosts (if all you have is a loopback
    >>> interface, only 127.0.0.1 will be needed),

    >
    >/etc/hosts
    >127.0.0.1 www.albury.net.au www localhost dxmm@albury.net.au
    >dxmm@albury dxmm
    >
    >I'm betting this is this why I cannot get to my ISP's home page -
    >www.albury.net.au


    Yup. And for what it's worth, the '@' character is not "legal" in a
    hostname. RFC0952 specifies 'caseless' (no difference between A and a),
    with letters, numbers, and the dash ONLY (name must begin with a letter
    and not end with a dash). Dots are used to divide domain/sub-domain/host
    only, and otherwise may not be part of the name. While RFC0952 is
    positively ancient (October 1985) and specified how hostnames were to
    be formed for use in the ARPA hosts file that pre-dated DNS, the same
    requirements exist in RFC1035 which covers DNS. The '@' sign is
    specified in RFC0821 as the separator between username and the domain
    in an email address. Using it in a hostname would get messy to say the
    least.

    > /etc/resolv.conf
    >
    >search net.au


    man 5 resolver The search (and domain) keywords are often more
    trouble than the keystrokes they save.

    ># search domain
    >nameserver 202.3.36.10
    >nameserver 202.3.36.11
    ># nameserver 127.0.0.1
    ># nameserver 127.0.0.1
    ># nameserver 127.0.0.1


    If you have control of this file, I'd delete the four lines shown here
    that begin with octathorpe (#) as unneeded. The 'nameserver 127.0.0.1'
    is redundant, as that is the default action (see the resolver man page).

    >removed www.albury.net.au from hosts file ... see how that goes!


    That really should fix things, because /etc/nsswitch.conf (and the
    older /etc/host.conf) usually specify to look in the hosts file first,
    and if the "answer" is found their, DNS won't be queried. Some use
    this concept as an ad-blocker (putting the hostname of the undesired
    ad server as 127.0.0.1) which as you've discovered blocks access to
    that crap. I prefer to use a firewall instead.

    >I haven't done this stuff that follows......little bits...see how I go!!


    [using a packet sniffer to see where your packets are getting lost]

    Shouldn't be needed. The hosts file entry was the reason you can't
    get to www.albury.net.au.

    Old guy

  18. Re: Using PPP to dial-in to ISP

    Moe Trin wrote:
    > On Mon, 03 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    > <47cbddb9$0$26047$88260bb3@free.teranews.com>, Daniel wrote:
    >
    >> Moe Trin wrote:

    >
    >>> Turn your debugging back on - and recheck the logs. Near the end of
    >>> the pppd man page, there is a section "EXIT STATUS" with the rather
    >>> simple-minded explanations of those codes.

    >
    >>> The peer didn't like what you said, and hung up the phone.

    >
    >> Yes, Moe, but this is not at my behest! I've suspected the line level
    >> may, at times, be marginal, and sometimes the line level falls below
    >> the modems minimum usable level, so the modem closes the connection.

    >
    > That _should_ only occur after the connection has been established,
    > as otherwise pppd would report a different exit code (probably 8).
    > Lessee, back in November, you reported this as a "Motorola Lifestyle"
    > modem, and the only such modem I'm aware of is an ancient 28.8k ISA
    > Internal modem that used a Motorola chip set of some kind.


    Motorola 28.8 Data/Fax modem, V.34 external desktop modem for personal
    computers

    That's what the front cover of manual says!

    Do you
    > have a manual for it? Most modems have a post-call diagnostics
    > command of some kind - unfortunately the command and data format is
    > not standardized across models from the same manufacturer, much less
    > from different ones. One of the things these commands can report is
    > the signal strength, and the signal to noise ratio. Another thing often
    > reported it the reason the link was terminated.
    >
    >> Is this far off the mark??

    >
    > It's certainly not impossible, but most modems will "work" with a pretty
    > bad signal level - something like 1/100 the level of a "normal" voice
    > connection. This would be VERY noticeable. Can you hear the modem
    > negotiations (the sounds the two modems make to each other when they
    > are trying to establish a connection)?
    >
    > Old guy


    Whilst I was looking through the Modem's handbook to see if I could find
    anything about post-call diags the connection died with this being the
    relevant part of the PPP-Logfile. Connected about 15:22, then at 16:45
    something decided to Hangup (Sighup). Any hints??

    Mar 4 15:12:16 www pppd[4285]: sent [PAP AuthReq id=0x2
    user="dxmm@albury.net.au" password=]
    Mar 4 15:12:16 www pppd[4285]: rcvd [PAP AuthAck id=0x2 ""]
    Mar 4 15:12:16 www pppd[4285]: PAP authentication succeeded
    Mar 4 15:12:16 www pppd[4285]: sent [CCP ConfReq id=0x1 -L -D -C> ]
    Mar 4 15:12:16 www pppd[4285]: sent [IPCP ConfReq id=0x1 0f 01> ]
    Mar 4 15:12:16 www pppd[4285]: rcvd [IPCP ConfReq id=0x7c 202.3.39.1>]
    Mar 4 15:12:16 www pppd[4285]: sent [IPCP ConfAck id=0x7c 202.3.39.1>]
    Mar 4 15:12:16 www pppd[4285]: rcvd [LCP ProtRej id=0x12 80 fd 01 01 00
    15 12 06 00 00 00 00 1a 04 78 00 18 04 78 00 15 03 2f]
    Mar 4 15:12:16 www pppd[4285]: rcvd [IPCP ConfRej id=0x1 0f 01>]
    Mar 4 15:12:16 www pppd[4285]: sent [IPCP ConfReq id=0x2
    ]
    Mar 4 15:12:17 www pppd[4285]: rcvd [IPCP ConfNak id=0x2 202.3.39.29> ]
    Mar 4 15:12:17 www pppd[4285]: sent [IPCP ConfReq id=0x3 202.3.39.29> ]
    Mar 4 15:12:17 www pppd[4285]: rcvd [IPCP ConfAck id=0x3 202.3.39.29> ]
    Mar 4 15:12:17 www pppd[4285]: local IP address 202.3.39.29
    Mar 4 15:12:17 www pppd[4285]: remote IP address 202.3.39.1
    Mar 4 15:12:17 www pppd[4285]: primary DNS address 202.3.36.10
    Mar 4 15:12:17 www pppd[4285]: secondary DNS address 202.3.36.11
    Mar 4 15:12:17 www pppd[4285]: Script /etc/ppp/ip-up started (pid 4323)
    Mar 4 15:12:22 www pppd[4285]: Script /etc/ppp/ip-up finished (pid
    4323), status = 0x0
    Mar 4 16:45:01 www pppd[4285]: Hangup (SIGHUP)
    Mar 4 16:45:01 www pppd[4285]: Modem hangup
    Mar 4 16:45:01 www pppd[4285]: Connect time 92.8 minutes.
    Mar 4 16:45:01 www pppd[4285]: Sent 653886 bytes, received 3274044 bytes.
    Mar 4 16:45:01 www pppd[4285]: Script /etc/ppp/ip-down started (pid 4777)
    Mar 4 16:45:01 www pppd[4285]: Connection terminated.
    Mar 4 16:45:01 www pppd[4285]: Waiting for 1 child processes...
    Mar 4 16:45:01 www pppd[4285]: script /etc/ppp/ip-down, pid 4777
    Mar 4 16:45:02 www pppd[4285]: Script /etc/ppp/ip-down finished (pid
    4777), status = 0x0
    Mar 4 16:45:02 www pppd[4285]: Exit.

    --
    Posted via a free Usenet account from http://www.teranews.com


  19. Re: Using PPP to dial-in to ISP

    On Tue, 04 Mar 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <47ccd825$0$26045$88260bb3@free.teranews.com>, Daniel wrote:

    >Moe Trin wrote:


    >> the only such modem I'm aware of is an ancient 28.8k ISA Internal
    >> modem that used a Motorola chip set of some kind.

    >
    >Motorola 28.8 Data/Fax modem, V.34 external desktop modem for personal
    >computers


    I suspect that is the same hardware in a different package.

    >That's what the front cover of manual says!


    Sounds good

    >Whilst I was looking through the Modem's handbook to see if I could find
    >anything about post-call diags the connection died with this being the
    >relevant part of the PPP-Logfile. Connected about 15:22, then at 16:45
    >something decided to Hangup (Sighup). Any hints??


    >Mar 4 15:12:16 www pppd[4285]: PAP authentication succeeded


    Still using PAP

    >Mar 4 15:12:16 www pppd[4285]: sent [CCP ConfReq id=0x1 >-L -D -C> ]


    You suggest using one of three flavors of Data Compression

    >Mar 4 15:12:16 www pppd[4285]: rcvd [LCP ProtRej id=0x12 80 fd 01 01 00
    >15 12 06 00 00 00 00 1a 04 78 00 18 04 78 00 15 03 2f]


    80fd Compression Control Protocol [RFC1962]

    The peer comes back and says "I don't do Compression" - You _could_ add
    the option "noccp" but it's really not needed as this was a clean
    rejection, and things are working normally.

    >Mar 4 15:12:22 www pppd[4285]: Script /etc/ppp/ip-up finished (pid
    >4323), status = 0x0
    >Mar 4 16:45:01 www pppd[4285]: Hangup (SIGHUP)
    >Mar 4 16:45:01 www pppd[4285]: Modem hangup
    >Mar 4 16:45:01 www pppd[4285]: Connect time 92.8 minutes.
    >Mar 4 16:45:01 www pppd[4285]: Sent 653886 bytes, received 3274044 bytes.


    Three possibilities: Modem/line problems (HIGHLY unlikely). ISP
    connection time limit - some ISPs restrict the amount of time you can
    be connected. And based on the number of bytes sent/received, an
    "Idle timeout" by your ISP, similar to the idle option in pppd:

    idle n Specifies that pppd should disconnect if the link is idle
    for n seconds. The link is idle when no data packets
    (i.e. IP packets) are being sent or received.

    There is also an LCP Echo timeout function, for which the pppd man page
    says

    lcp-echo-failure n
    If this option is given, pppd will presume the peer to be
    dead if n LCP echo-requests are sent without receiving a
    valid LCP echo-reply. If this happens, pppd will termi-
    nate the connection. Use of this option requires a non-
    zero value for the lcp-echo-interval parameter.

    Now I'm pretty sure you are not using this option, and I don't see an
    indication that the peer is doing so either. I mention this only as a
    possibility. Your link would have to degrade pretty badly before this
    would be a concern. A quick check would be to look at the /sbin/ifconfig
    output after you've been connected for some time - is the ppp section
    showing any error counts above zero?

    Old guy

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2