Date: Thu, 10 Aug 1995 23:47:25 -0400 (EDT) From: Jim Durham <durham@w2xo.pgh.pa.us> To: hackers@freebsd.org Subject: Network behavior if connection fails. Message-ID: <Pine.BSF.3.91.950810233630.2995C-100000@w2xo.pgh.pa.us>
next in thread | raw e-mail | index | archive | help
I have noticed a phenomenum with both 1.1.5.1 and 2.0.5 . This may
be normal behavior, but it's difficult to deal with.
If the network connection goes down somewhere beyond the PPP connection,
then the whole networking system internally locks up.
When this happens, I can't telnet to localhost and I can't log in
on my local ethernet from my "other" machine to the affected machine.
This isn't the name-server waiting to time-out. This doesn't seem to
clear itself. The only way I can get things working again is to
blow away the PPP process, then run Kermit on the port and do "hangup"
to hang up the modem, then restart PPP (I'm running it as a daemon, using
iij PPP).
My internet provider has been having touble with his T1 line, so
I've had this happen several times. Strangely enough, the internet
provider for my machine at work had his Livingston router go bonkers
and , although it would allow a SLIP login, it wouldn't pass packets.
This caused *that* machine to do the same thing.
Is this normal behavior? It's really a drag. If you've got a good, solid
connection, everything is fine, but if you start having trouble with the net,
everything just stops. Yes, I do have named running...
I also noticed that it takes about 2 minutes to log in or su to root
under this condition. Why does 'su" care about the network?
A puzzled...
-Jim Durham
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.950810233630.2995C-100000>