While Roaring Penquin pppoe has a nice tkpppoe interface that displays
transfer rates, I just noticed that when downloading an iso file, I was
getting more errors than RX packets and erratic speeds. But with no
drops, overruns, frame errors or collisions, the nature of the errors is
beyond me. I also noticed a server load of .3, but that may have been due
to animated images on the page the download link was on.
When I use kernel pppoe (set up with YaST2 and fired up with kinternet,
transfer is very steady (judging from ncftp) and no errors at all (and no
server load).
Examples (snipped):
rp-pppoe using tkpppoe (larger download exhibited same symptoms):
ppp0 Link encap:Point-to-Point Protocol
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492 Metric:1
RX packets:1992 errors:2004 dropped:0 overruns:0 frame:0
TX packets:2000 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3
RX bytes:2769193 (2.6 Mb) TX bytes:104402 (101.9 Kb)
Kernel pppoe started from kinternet:
ppp0 Link encap:Point-to-Point Protocol
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492 Metric:1
RX packets:474853 errors:0 dropped:0 overruns:0 frame:0
TX packets:470639 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:3
RX bytes:700757844 (668.2 Mb) TX bytes:24487877 (23.3 Mb)
Anyone else noticed this difference between rp-pppoe and pppoe? Any idea
if it is caused by the HDLC that rp-pppoe uses or something related to the
graphic display in ktpppoe?
--
David Efflandt - All spam is ignored - http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/