I long ago noticed that if I ping somewhere on the internet for any
length of time I see a pattern - the ping times will slowly rise then
fall by 10ms and repeat the process.
I never asked before because my dsl modem taints the kernel - but I
think testing I did on 2.4s could mean that it's not the issue.
I post now because there is a thread on a UK forum where it has been
noticed that linux kit + PCI modems/ routers running Linux give worse
pings than Windows + PCI or other router/modems.
The reason I (possibly erroneously) think that the modem is not the
issue is that one test (on 2.4 long ago) I did involved pinging my first
hop simoultaneously from the gateway and a lan PC.
The gateway was always + 10ms the forwarded traffic followed the pattern
of slowly rising then falling by 10ms and repeating - always.
After changing kernels it changed so that the forwarded pings would fall
then rise.
The 10ms pointed to hz 100 - so I patched and ran 500, sure enough the
varience was now 2ms - sorted I thought.
Now on 2.6 with Hz 1000 it is back to 10ms rather than 1ms. It's not
quite the same - a bit more random and both forwarded and local cycle.
The modem is a connexent access runner drivers from
http://patrick.spacesurfer.com/linux_conexant_pci_adsl.html
2.4 http://patrick.spacesurfer.com/adsl/CnxADSL-6.1.2.007-PIM-4.tar.bz2
2.6 http://patrick.spacesurfer.com/adsl/driver-2.6-latest.tar.bz2
The connection to the internet is pppoa. Pinging the lan card on the
gateway gives a steady 0.3ms.
Below is an example - traffic shaping turned off for the test.
Does anyone know what could be causing this.
TIA
Andy.
64 bytes from 82.133.85.65: icmp_seq=22 ttl=58 time=22.157 ms
64 bytes from 82.133.85.65: icmp_seq=23 ttl=58 time=24.497 ms
64 bytes from 82.133.85.65: icmp_seq=24 ttl=58 time=24.833 ms
64 bytes from 82.133.85.65: icmp_seq=25 ttl=58 time=23.947 ms
64 bytes from 82.133.85.65: icmp_seq=26 ttl=58 time=22.327 ms
64 bytes from 82.133.85.65: icmp_seq=27 ttl=58 time=22.696 ms
64 bytes from 82.133.85.65: icmp_seq=28 ttl=58 time=22.042 ms
64 bytes from 82.133.85.65: icmp_seq=29 ttl=58 time=22.371 ms
64 bytes from 82.133.85.65: icmp_seq=30 ttl=58 time=22.667 ms
64 bytes from 82.133.85.65: icmp_seq=31 ttl=58 time=23.959 ms
64 bytes from 82.133.85.65: icmp_seq=32 ttl=58 time=24.264 ms
64 bytes from 82.133.85.65: icmp_seq=33 ttl=58 time=23.570 ms
64 bytes from 82.133.85.65: icmp_seq=34 ttl=58 time=23.895 ms
64 bytes from 82.133.85.65: icmp_seq=35 ttl=58 time=13.123 ms
64 bytes from 82.133.85.65: icmp_seq=36 ttl=58 time=13.448 ms
64 bytes from 82.133.85.65: icmp_seq=37 ttl=58 time=14.774 ms
64 bytes from 82.133.85.65: icmp_seq=38 ttl=58 time=15.069 ms
64 bytes from 82.133.85.65: icmp_seq=39 ttl=58 time=15.392 ms
64 bytes from 82.133.85.65: icmp_seq=40 ttl=58 time=14.672 ms
64 bytes from 82.133.85.65: icmp_seq=41 ttl=58 time=14.995 ms
64 bytes from 82.133.85.65: icmp_seq=42 ttl=58 time=16.289 ms
64 bytes from 82.133.85.65: icmp_seq=43 ttl=58 time=14.628 ms
64 bytes from 82.133.85.65: icmp_seq=44 ttl=58 time=14.932 ms
64 bytes from 82.133.85.65: icmp_seq=45 ttl=58 time=16.218 ms
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
