quick question

[kancer at kancer.978.org (Kris)]

Yeah I have 2 linux boxs on my small home network over a 100mb switch and
I get the same thing when logged into both pinging each other.  
I just figured it to be the ethernet backoff algorithm in effect.
Or different priorities on the switch.
Or just that some network cards have higher RX latency than TX latency.
Or system time is skewed on one box not the other.
Basically everything is working the way it was engineered 20+ years ago.
When you see anything over 10ms, errors and/or dropped packets is when
there is an issue.


Kris Loranger
kris at kancer.978.org
IRC:undernet,#978,Kancer AIM:KancerKris
"If you're going to sell out, sell out ethically" -Moby

On 13 Feb 2001, Derek Atkins wrote:

> Keep in mind that your timers may be of different resolution.  For
> example, your "other machine" may have 1ms granularity, whereas your
> linux box as 0.1ms granularity.  I would be surprised if you saw 0ms
> in one direction and 3-4ms in the other.  But you're seeing '0' and
> '0.6' which are basically equivalent if the former has a 1ms
> granularity.
> 
> In other words, anything <1ms appears as 0ms, and obviously 0.6 < 1
> 
> -derek
> 
> Nick Oleksinski <noleks at lsil.com> writes:
> 
> > Has anyone ever seen a situation where outgoing pings from a machine are
> > slow on a LAN but ping reflections (ie another machine on the LAN
> > pinging the Linux box) are very fast?  I'm getting this on a Linux box
> > using a 3Com 3C905C-TX ethernet card and kernel 2.4.1.  The times look
> > something like this:
> > 
> > FROM ANOTHER MACHINE TO THE PC (wmpc75):
> > noleks tcsh[2]$ ping -s wmpc75
> > PING wmpc75: 56 data bytes
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=0. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=1. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=2. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=3. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=4. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=5. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=6. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=7. time=0. ms
> > 64 bytes from wmpc75 (147.145.153.37): icmp_seq=8. time=0. ms
> > ^C
> > ----wmpc75 PING Statistics----
> > 9 packets transmitted, 9 packets received, 0% packet loss
> > round-trip (ms)  min/avg/max = 0/0/0
> > noleks tcsh[3]$ 
> > 
> > 
> > TO ANOTHER MACHINE FROM THE PC (wmpc75):
> > noleks tcsh[2]$ ping wm27
> > PING wm27 (147.145.152.27): 56 octets data
> > 64 octets from 147.145.152.27: icmp_seq=0 ttl=254 time=1.8 ms
> > 64 octets from 147.145.152.27: icmp_seq=1 ttl=254 time=0.6 ms
> > 64 octets from 147.145.152.27: icmp_seq=2 ttl=254 time=0.6 ms
> > 64 octets from 147.145.152.27: icmp_seq=3 ttl=254 time=0.6 ms
> > 64 octets from 147.145.152.27: icmp_seq=4 ttl=254 time=0.6 ms
> > 64 octets from 147.145.152.27: icmp_seq=5 ttl=254 time=0.6 ms
> > 64 octets from 147.145.152.27: icmp_seq=6 ttl=254 time=0.7 ms
> > 
> > --- wm27 ping statistics ---
> > 7 packets transmitted, 7 packets received, 0% packet loss
> > round-trip min/avg/max = 0.6/0.7/1.8 ms
> > noleks tcsh[3]$ 
> > 
> > Thanks
> > -Nick
> > -
> > Subcription/unsubscription/info requests: send e-mail with
> > "subscribe", "unsubscribe", or "info" on the first line of the
> > message body to discuss-request at blu.org (Subject line is ignored).
> 
> -- 
>        Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
>        Member, MIT Student Information Processing Board  (SIPB)
>        URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
>        warlord at MIT.EDU                        PGP key available
> -
> Subcription/unsubscription/info requests: send e-mail with
> "subscribe", "unsubscribe", or "info" on the first line of the
> message body to discuss-request at blu.org (Subject line is ignored).
> 

-
Subcription/unsubscription/info requests: send e-mail with
"subscribe", "unsubscribe", or "info" on the first line of the
message body to discuss-request at blu.org (Subject line is ignored).