[On Topic] Re: Dial in & out - Fixed. (doh!)

• Messages sorted by:[ date ][ thread ][ subject ][ author ]

In article <19991117225308.A28803@greenie.muc.de>, gert@greenie.muc.de says...
>Hi,
>
>On Wed, Nov 17, 1999 at 05:31:53PM +0000, mgetty@nexnix.co.uk wrote:
>> > annoy all the ISPs that permit spamming, etc.
>> 
>> Yup, I'm quite practised at that.
>> I do like to get the spammers home phone number as well,
>> surprisingly easy for the amateurs.
>
>ok, ok, you've won :)
"whois" is your friend ;)

Thanks for the reply. Sorry for the tardy reply - I had to go to site.


>
>Most likely the problem with dial-in/dial-out not working is caused by
>missing/wrong lock files.  Mgetty can only "go away" peacefully when a
>dialout is running if it can detect a "LCK..ttyS0" file telling it "the
>tty is mine" (otherwise mgetty has to assume an incoming call).

OK I can grok that. So it looks like mgetty is not detecting the LCK..*
correctly.  Or something.

Pppd is making lockfiles in /var/lock (LCK..ttyS1) but from the process
table mgetty and pppd are both running at the same time. Normal?


Ah. No it's not. mgetty is making LCK..ttyS1 - pppd is making LCK..modem.
So the lesson learnt is to make sure that the device is called by the same
name (even if it's a link) throughout all the applications that require
lockfiles.  Obvious really when you think about it. ;>

Once I told mgetty to use /dev/modem (the same as pppd) the locking
mechanism works properly and I can dialout while mgetty is running
because mgetty sleeps on pppds lockfile. Yay!


Thanks for the help, it's been a pleasure talking with you.
And thanks again for mgetty, without which I would probably
have to use NT + RAS <ack>


ttfn
mgetty@nexnix (R.I.P)

(I guess I should post this on the newsgroup so perhaps it'll find it's
way into DejaNews for the next one who makes the same mistake)

Oh. I have.  :)