Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 22 Apr 1997 00:32:42 +0300 (EET DST)
From:      mika ruohotie <bsdcur@shadows.aeon.net>
To:        current@freebsd.org
Subject:   dup and other anomalities...
Message-ID:  <199704212132.AAA08722@shadows.aeon.net>

next in thread | raw e-mail | index | archive | help
well, is there something wrong, or am i missing something?

my friend has weird problem in his net, check these pings (this is /27),
the problem introduced himself after he upgraded from 2.1.5 to 2.2.-stable
with make world. before in 2.1.5 no probs.

(this is actually real c-class address, but he idnt want it to be visible)

10.0.2.17 -> 10.0.2.18

PING 10.0.2.18 (10.0.2.18): 56 data bytes
64 bytes from 10.0.2.18: icmp_seq=0 ttl=255 time=90.152 ms
64 bytes from 10.0.2.18: icmp_seq=1 ttl=255 time=79.985 ms
64 bytes from 10.0.2.18: icmp_seq=2 ttl=255 time=69.991 ms
64 bytes from 10.0.2.18: icmp_seq=3 ttl=255 time=59.990 ms
64 bytes from 10.0.2.18: icmp_seq=4 ttl=255 time=49.972 ms
64 bytes from 10.0.2.18: icmp_seq=5 ttl=255 time=39.990 ms
64 bytes from 10.0.2.18: icmp_seq=6 ttl=255 time=29.990 ms
64 bytes from 10.0.2.18: icmp_seq=7 ttl=255 time=19.992 ms
64 bytes from 10.0.2.18: icmp_seq=8 ttl=255 time=9.987 ms

the cycle goes on and on, doesnt always start from 90.

then:

10.0.2.17 -> netmask 10.0.2.31

PING 10.0.2.31 (10.0.2.31): 56 data bytes                               
64 bytes from 10.0.2.17: icmp_seq=0 ttl=255 time=0.371 ms                   
64 bytes from 10.0.2.18: icmp_seq=0 ttl=255 time=90.442 ms (DUP!)           
64 bytes from 10.0.2.17: icmp_seq=1 ttl=255 time=0.299 ms                   
64 bytes from 10.0.2.18: icmp_seq=1 ttl=255 time=90.032 ms (DUP!)           
64 bytes from 10.0.2.17: icmp_seq=2 ttl=255 time=0.287 ms                   
64 bytes from 10.0.2.18: icmp_seq=2 ttl=255 time=80.044 ms (DUP!)           
64 bytes from 10.0.2.17: icmp_seq=3 ttl=255 time=0.289 ms                   
64 bytes from 10.0.2.18: icmp_seq=3 ttl=255 time=70.061 ms (DUP!)           
64 bytes from 10.0.2.17: icmp_seq=4 ttl=255 time=0.285 ms                   
64 bytes from 10.0.2.18: icmp_seq=4 ttl=255 time=60.044 ms (DUP!)

no routing daemon, static routes.

with routed that worked in 2.1.5 nicely, there's this:

Apr 20 12:55:26 soapbar /kernel: ipfw: 2299 Allow UDP 10.0.2.17:520 
10.0.2.31:520 via de0

what the broadcast does there?

then, i did this same in my isp's 2.2 machine: (pinged the broadcast)
(i hide the real ip coz they probably dont want it out, it's c-class too)

PING 10.0.0.127 (10.0.0.127): 56 data bytes
64 bytes from 10.0.0.15: icmp_seq=0 ttl=255 time=0.547 ms
64 bytes from 10.0.0.1: icmp_seq=0 ttl=255 time=20.393 ms (DUP!)
64 bytes from 10.0.0.11: icmp_seq=0 ttl=255 time=21.575 ms (DUP!)
64 bytes from 10.0.0.111: icmp_seq=0 ttl=255 time=21.775 ms (DUP!)
64 bytes from 10.0.0.3: icmp_seq=0 ttl=255 time=28.973 ms (DUP!)
64 bytes from 10.0.0.4: icmp_seq=0 ttl=255 time=49.307 ms (DUP!)
64 bytes from 10.0.0.126: icmp_seq=0 ttl=255 time=156.040 ms (DUP!)
64 bytes from 10.0.0.5: icmp_seq=0 ttl=255 time=422.538 ms (DUP!)

others than 126 are frebsd 2.2 machines (as far as i know) and 126 is cisco.

as far as i can tell again, my isp has static routes too.

now, i cant generate that in either 2.1.-stable net at work, nor in my
3.0-current net at home... so i assume there's something broken, i
can not really say what, in the 2.2, right? if not, what's wrong?

another question, why is it possible to ping the broadcast anyway?

also, now that i think about it, a little after my isp went to 2.2 i started
to notice some lag, but it depends... (this is a prob my pal has in his net
too, and i might have similar prob too)

like, freebsd-freebsd connection goes fine, but freebsd-cisco doesnt,
then again freebsd-freebsd-cisco goes fine... (we are talking about 2.2 here)

and from my net to my isp's 2.2 machine running ircd, without a reason
there is a severe lag specially from the win95 box. as far as i can tell
my freebsd 3.0 works fine with it. also as far as i can tell, i cant trace
from where the lag comes, it just _is_ there... (traceroute/ping look clean)

my hunch would be something bad in the tcp section, but is it so? is
_anyone_ else having anything even closely related to this? specially
eyes on 2.2-branch. (i personally dont run it yet on anything, not until
i get my release cd, after which i upgrade 2.1 to it and cvsup that to 
2.2-latest)

(and if i sound like i'm not making sense, it's after midnite, pardon me)


mickey



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704212132.AAA08722>