I have run the command 'tcpdump -w tcpdump.txt | yum update' and then i used
'tcpdump -r tcpdump.txt' to get the results, below:
reading from file tcpdump.txt, link-type EN10MB (Ethernet)
12:39:09.221619 arp who-has mygateway.ar7 tell sicoserver
12:39:09.225138 arp reply mygateway.ar7 is-at 00:14:6c:10:6e:da (oui
Unknown)
12:39:09.225145 IP sicoserver.35396 > mygateway.ar7.domain: 36584+ AAAA?
mirror.bytemark.co.uk. (39)
12:39:09.267061 IP mygateway.ar7.domain > sicoserver.35396: 36584 0/0/0
(39)
12:39:09.267140 IP sicoserver.39452 > mygateway.ar7.domain: 9735+ AAAA?
mirror.bytemark.co.uk. (39)
12:39:09.272099 IP mygateway.ar7.domain > sicoserver.39452: 9735- 1/0/0
(39)
12:39:09.272166 IP sicoserver.35949 > mygateway.ar7.domain: 9418+ A?
mirror.bytemark.co.uk. (39)
12:39:09.276625 IP mygateway.ar7.domain > sicoserver.35949: 9418- 1/0/0
A[|domain]
12:39:09.276778 IP sicoserver.36733 > 1.0.0.0.http: S 509686349:509686349(0)
win 5840 <mss 1460,sackOK,timestamp 220963 0,nop,wscale 7>
12:39:12.276614 IP sicoserver.36733 > 1.0.0.0.http: S 509686349:509686349(0)
win 5840 <mss 1460,sackOK,timestamp 223963 0,nop,wscale 7>
12:39:14.256873 arp who-has sicoserver tell mygateway.ar7
12:39:14.256880 arp reply sicoserver is-at 00:18:f3:67:f7:fe (oui Unknown)
12:39:18.276669 IP sicoserver.36733 > 1.0.0.0.http: S 509686349:509686349(0)
win 5840 <mss 1460,sackOK,timestamp 229963 0,nop,wscale 7>
12:39:28.475768 IP 192.168.0.7.plysrv-https > 239.192.152.143.plysrv-https:
UDP, length 119
12:39:30.276781 IP sicoserver.36733 > 1.0.0.0.http: S 509686349:509686349(0)
win 5840 <mss 1460,sackOK,timestamp 241963 0,nop,wscale 7>
12:39:35.276821 arp who-has mygateway.ar7 tell sicoserver
12:39:35.278773 arp reply mygateway.ar7 is-at 00:14:6c:10:6e:da (oui
Unknown)
12:40:10.367175 IP 192.168.0.7.netbios-dgm > 192.168.0.255.netbios-dgm: NBT
UDP PACKET(138)
A lot of this doesn't make sense to me, but I can work out that packets are
being sent to and from my server. What looks strange to me though is that
my server has tried to send packets to "1.0.0.0.http"? Should this not be a
valid IP address?
My 'yum.conf' file has no reference to neither throttle nor bandwidth. Here
is the contents of my file, incase there is something wrong with it:
[main]
cachedir=/var/cache/yum
keepcache=0
debuglevel=2
logfile=/var/log/yum.log
distroverpkg=redhat-release
tolerant=1
exactarch=1
obsoletes=1
gpgcheck=1
plugins=1
# Note: yum-RHN-plugin doesn't honor this.
metadata_expire=1h
installonly_limit = 5
# PUT YOUR REPOS HERE OR IN separate files named file.repo
# in /etc/yum.repos.d
This is really annoying me now, it seems to me like I have perfectly good
internet connection (pings and curls work fine) and I can access the needed
file from other computers on the network but it just wont get the file when
running yum.
--------------------------------------------------
From: "Andrew Vliet" <arvliet@xxxxxxxxx>
Sent: Sunday, September 13, 2009 9:56 PM
To: "Yellowdog Updater, Modified" <yum@xxxxxxxxxxxxxxxxx>
Subject: Re: 'yum update' error - Cannot retrieve repository metadata
On Sun, 2009-09-13 at 19:24 +0100, Andrew Wheal wrote:
After a fresh install of CentOS 5.3, 'yum update' has been been
failing to run. I have posted this problem on the CentOS forums,
which can be found at this link:
https://www.centos.org/modules/newbb/viewtopic.php?topic_id=22178
The people on that forum have been very helpful so far but have run
out of ideas why it is not working. One member suggested that I come
here to see if anyone is able to help.
When I run 'yum update' now the output is as follows:
Loaded plugins: fastestmirror
Determining fastest mirrors
http://mirror.bytemark.co.uk/centos/5/os/x86_64/repodata/repomd.xml:
[Errno 12] Timeout: <urlopen error timed out>
Trying other mirror.
Error: Cannot retrieve repository metadata (repomd.xml) for
repository: base. Please verify its path and try again
Any help with this problem would be much appreciated.
_______________________________________________
Yum mailing list
Yum@xxxxxxxxxxxxxxxxx
http://lists.baseurl.org/mailman/listinfo/yum
I actually joined this list trying to find an answer very similar
symptoms. It looked like things were working - but it took 13 minutes
to download the Fedora repo repomd.xml file. Was going to take an
estimate 15 thousand hours to download the primary_sqlite database. Uh
- my connection's bad, but not that bad... It turns out that I forgot
to put a "k" behind my bandwidth argument in my yum.conf. So, my
connection was 1024 times slower than I expected, but ironically, 100%
functionally accurate for what I'd told it to do... DOH!
Broken:
throttle=60%
bandwidth=12.5
Correct:
throttle=60%
bandwidth=12.5k
You can also run a tcpdump to make sure traffic is actually passing
to/from your computer when you do the yum update.
_______________________________________________
Yum mailing list
Yum@xxxxxxxxxxxxxxxxx
http://lists.baseurl.org/mailman/listinfo/yum
_______________________________________________
Yum mailing list
Yum@xxxxxxxxxxxxxxxxx
http://lists.baseurl.org/mailman/listinfo/yum
