Odd tcp dump? was: ssh working with dialup, not through router

M. Fioretti mfioretti at mclink.it
Sat May 15 21:51:31 UTC 2004


Hello,

still trying to track why ssh doesn't work anymore, see my original
posting.

the PC is192.168.1.2 and the ADSL router 192.168.1.1. I have run
tcpdump on the PC eth0 interface, while the ssh connection
freezes, and found that, at a certain point:

1) the PC asks the DNS server (if I understand correctly) about the
router:

192.168.1.2.1047 > ammi.mclink.it.domain:  22723+ PTR? 1.1.168.192.in-addr.arpa. (42) (DF)
ammi.mclink.it.domain > 192.168.1.2.1047:  22723 NXDomain 0/1/0 (119)

2) after that PC and router start to ask each other their ethernet
addresses ("arp who-has 192.168.1.2 tell 192.168.1.1" and the related
arp replies) and viceversa, repeatedly.

3) when this happens, the pc and the ssh server seem to also go in a
loop, which eventually times out: several equal lines in the dump,
like:

23:27:10.780631 192.168.1.2.1772 > ssh.server.io.ssh: . ack 2096 win 8832 <nop,nop,timestamp 474677 141420442,nop,nop,sack sack 1 {2048:2096} > (DF) [tos 0x10] 

Is 3) caused by 1) and 2)?

are 1) and 2) normal?

if they are not normal and causing ssh to freeze, how to fix them?

TIA,
	Marco F.

-- 
Marco Fioretti                 m.fioretti, at the server inwind.it
Red Hat for low memory         http://www.rule-project.org/en/

In a [wired by the Internet] mass media world, there is *less*  of
everything except the top ten books, records, movies, ideas.
                               -- Michael Crichton, "The Lost World"





More information about the fedora-list mailing list