Until yesterday I've used kernel 2.0.30 for my linux box (PPP
server with 4x16550, 2x 3c509).
The 2.0.30 kernel used to lock up on TCP connections from the
Linux side to the SVR4.2. _Very_ often. When I changed to another
VC and made a ping(1) to the SVR4.2 box, the connection awakened
again. And making a TCP connection (TELNET) from the SVR4.2 box
to the Linux machine had also no problems.
Under pre-2.0.31 it seemed that this problem has gone. Until now.
Connections to our PPP clients lock up from time to time. It seems
that connections to the SVR4.2 have no problems. I can't say that for
sure, however. It seems that connections tend to lock-up when the
PPP client tries to transfer a large amount of data (a big file
with FTP while making other short-living connections, like
playing around with www. The lockups happen on any PPP modem
line and also with only one client connected.
Because the TCP stalls under 2.0.30 and pre-2.0.31 happen
between both the SVR4.2 _and_ the PPP clients, it can't be a
network hardware related problem. And our ethernet is currently
very small: It contains only the SVR4.2 and the Linux box.
o Has anybody made similar experiences?
o Is there a known bugfix for this?
o Could it be a problem with ARP and/or proxy ARP? I wonder about
the 'ping' effect?
o If I can help tracing this down, tell me (how).
For now I try to get more information using tcpdump.
pm