According to Arno Verhoeven: While burning my CPU.
>
> Hi,
>
> I realize that this is going to be a long mail, and is therefor boring
> to read :^) But I hope you will make it to the nd of this mail and
> hopefully have a solution for my problem.
>
> For quite some time now we are havning problems with periodic
> disconnects.
>
> Until recently we had a PC equiped with a SCC card. But the SCC port for
> the full duplex link with the neighboring node locked up several times
> per day.
Ronnie Hale pointed out your disconnect problem, i would like to comment
on the SCC problem, please note i said "the" SCC problem and not "your" SCC
problem as it seems you are by NO means alone.
Over a period of time, there has been many comments about SCC interfaces
dissapiering and just sitting there doing only half the work, by that i
mean only receiving data but refusing to transmit.
I would be interested to know more about your problem as we at PI1GOE have
a problem with one interface doing as per described above all the time.
By that i mean also 2@3 times a day.
Our Symptoms are,
1) data is received on a 9k6(scc) interface (which never dies i might add)
and is retransmitted on a 4k8 interface.
2) Multiple sendmail sockets open (3@4 at a time to one host) on the 4k8
interface.
It does not need to be sendmail, its just an example, when multiple hosts
are ftping to, or via PI1GOE the same problem arises, the same applies to
HTTP connections etc.
It would seem buffers get overloaded, or something to that effect, ie, the
9k6 interface works twice as fast as the data can be retransimited on the
4k8 interface causing buffer/memory problems. Having sat behind different
"listen" screens for hours at a time by PI1GOE i get the impression the
problem is caused by the 9k6 interface acking packets like billyo and not
having any room to store them as the queue's get higher and higher on the
4k8 interface, with the enevatable effect that the 4k8 interface just stops
transmitting, when that happens, it sees every packet on the 4k8 interface
but just refuses to transmit.
There is no need to reboot to get over the problem a small script i have
running via cron, checks the interface with pings every 15 minutes, no pings
returned will cause ifconfig to restart the interface and all is well again.
There are never any signs or messages anywhere which could give even the
smallest hint of where to start looking for the problem.
I have explaned this problem to this list at least TWICE in the past, it is
NOT only me that has this problem, a look at the archives will produce quite
a few callsigns who all say they have the same symptoms in one way or
another.
Joreg DK1BKE has never responded to any of the comments, hints or kinks,
made about this problem, i have tryed many things different timings, MTU's,
window sizes, buffsize in /etc/z8530drv.conf, parms settings, realy you name
it i've tryed it, all to no avail, there is a problem with the SCC driver
but as finding out what it is, is like banging your head against a wall.
Arno, if you would like a copy of my script which checks the interface
activity with pings let me know, i will send you a copy, be warned its
system spesific, however i am sure you will be able to convert it to your
needs.
And you thought your mail was long!!!!!.
If there are others with this problem please speak now.
>
> Thanks in advance.
>
> 73, Arno pe1icq
--
Regards Richard.
[EMAIL PROTECTED]
