Date: Thu, 23 Feb 2006 23:47:37 +0100 From: "antonio montagnani" antonio.montagnani@gmail.com Subject: Re: Wget, Yum and network investigation To: "For users of Fedora Core releases" fedora-list@redhat.com Message-ID: 4c37b6af0602231447p44562489t@mail.gmail.com Content-Type: text/plain; charset=ISO-8859-1
2006/2/23, Tod Merley todbot88@gmail.com:
Date: Thu, 23 Feb 2006 18:27:20 +0100 From: "antonio montagnani" < antonio.montagnani@gmail.com> Subject: Re: Wget, Yum and network investigation To: "For users of Fedora Core releases" fedora-list@redhat.com Message-ID: < 4c37b6af0602230927w6098941k@mail.gmail.com> Content-Type: text/plain; charset=UTF-8
Hi Antonio!
From the thread I do not know which of the routers you updated from the
RH8,
but I would guess that it is the one you are having problems on. That
is
just a guess. It may be that something related to the IPv6 stack
handling
was not handled in the process.
Yes...it is the updated router: but it worked for a long time with FC5 and I didn't have any problem at all with yum, wget....I can't understand what changed
I would be most interested in the contents of /etc/resolv.conf on all machines. It would be nice to know who is being looked at for name resolution.
I will post the troubled router tomorrow morning as now I am at home. Anyway this is the resolv.conf of this machine, i.e. the router with no problems....
nameserver 62.211.69.150 nameserver 212.48.4.15
and modprobe.conf
alias eth0 ne2k-pci alias eth1 sk98lin alias eth2 hisax alias snd-card-0 snd-intel8x0 options snd-card-0 index=0 install snd-intel8x0 /sbin/modprobe --ignore-install snd-intel8x0 && /usr/sbin/alsactl restore >/dev/null 2>&1 || : remove snd-intel8x0 { /usr/sbin/alsactl store >/dev/null 2>&1 || : ; }; /sbin/modprobe -r --ignore-remove snd-intel8x0 alias usb-controller ehci-hcd alias usb-controller1 uhci-hcd
When I troubleshot an IPv6 name resolution problem at home here I used "tcpdump -w captureFileName &" along with Ethereal to analyze the
tcpdump
capture files. When I did it some of the packets were truncated so it
would
be best to use -s 0 (capture packets of arbitrary length) or -s
1515(capture
packets as large as the max Ethernet frame) in the tcpdump command.
In my case, turning off IPv6 (accomplished, I believe, by adding "alias net-pf-10 off" to /etc/modprobe.conf and rebooting) did resolve the
problem
on a single FC4 machine. Since I had an Ubuntu machine on the same
network
and could see no way to effectively turn off IPv6 on that machine I
simply
routed nameservice arround the DSL modem which appeared to have problems with IPv6 name serving (probably a frimware problem) and the problem
went
away. Of course to do this there needed to be an alternative nameserver
in
/etc/resolv.conf.
Ipv6 has been turned off on office router, but no improvement.What went wrong??
Tod
-- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
-- Antonio Skype : antoniomontag
Hi again Antonio!
I just have a bit of a thought here. You may have simply a problem with the URL names requested by yum. Is it possible that somehow the repositories have been re-named (their URLs) and that was not completed in the update process. If I try to nslookup or whois the long repo URLs you sent in your first message they fail. Are the addresses differant (the requested URL's yum is attempting to access) on the two machines? If I point firefox at the addresses it finds them - perhaps the "redirection" switch in a yum config file is set differently. Just bits of thoughts.
I must say I am curious what kind of Internet access you have over there? How do you get Internet to the machines?
Ethereal gets down to the nitty gritty and would probably be a good one to do here. Of course it takes time.
Good Hunting!
Tod