Problem with ncpfs since 2.3.49. Maybe networking?

From: Petr Vandrovec (VANDROVE@vc.cvut.cz)
Date: Thu Mar 16 2000 - 17:10:34 EST


Hi,
  I have problems with ncpfs since 2.3.49. If I do
ls ~mirror/remote (it is ncpfs directory) under anything
below or equal to 2.3.49-pre1, it takes 0.077s or real time. If
I do it under 2.3.49 (or laters including 2.3.99-pre1), it takes
either 2.554s or 1.940s.
  When I look to network traffic, there is simple one packet
missing, and after about 500ms ncpfs does retransimt. And because
it happens 4-5 times during that directory listing ... :-(
  So my question is: Is it known problem that either tulip
driver or networking itself (it happens on both IPX and UDP)
has problems in transmit and drop frames instead of transmitting
them? There is nothing in the logs and during tests machine
has 220MB free (+32MB used + 15MB cached), so ENOMEM should not
occur.
  Networking card is 1011:0019 (rev 30), subsystem 1113:1207.
  Machine is dual PIII/450, 256MB RAM. Config file is available
at ftp://vana.vc.cvut.cz/pub/machine. Machine itself is
vana.vc.cvut.cz, if you are interested. But 'ping -f' does not
reveal any problems. I can try to recompile kernel without
IP multicasting and filtering, if it can cause these troubles.
                        Thanks,
                                        Petr Vandrovec
                                        vandrove@vc.cvut.cz
                                        
P.S.: I think that problem is 2.4.0 showstopper :-(

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Thu Mar 23 2000 - 21:00:18 EST