traceroute error !<10>

• Previous message: Phil Schaffner: "Re: Need Octave Forge rpm"
• In reply to: Phil Schaffner: "Re: special install situation"
• Next in thread: Alexander Dalloz: "Re: traceroute error !<10>"
• Reply: Alexander Dalloz: "Re: traceroute error !<10>"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

To: For users of Fedora Core releases <fedora-list@redhat.com>
Date: Sun, 28 Nov 2004 14:12:19 +1000

Sorry for this long message, but I'm hopping someone will know
exactly what is happening with the error, or atleast point me in the
right direction. Thanks.

I'm working on upgrading my one lab of 20 computers to FC3, and
have run into a strange problem. Got around bugzilla problems
138419, and 139724 already, and am now able to get FC3 to install
on the machines with 98 and XP.

I updated on machine yesterday, and was doing another today, but
rather than redownloading the updates again, I was just going to
transfer over the contents of the /var/cache/apt/archives directory. I
archived the contents into a tar.gz file, and ftped to the machine.
Logged in, but a directory comes back with no route to host. Here is
the strange part. Both machines are plugged into the same hub,
traceroute shows a response in about 0.200ms, but then has the
!<10> after it. What is also weird, is both machines have no problem
ftping to an old windows 98 machine, so I ended up ftping the file to
the windows 98 machine on another subnet, and then downloading
it to the other machine.

Not clear on what this !<10> error is with router solicitation?

After untarring the file, the update went directly to installing all the
files, so that went fine.

I had a similar problem about a week ago. I have FC2 on the 20
machines, and was doing a synaptic update of all the machine,
which use another machine with squid to cache updates. 18
machines went thru just fine, but two machines couldn't see the
squid machine at all. Had to disable the proxy and have them
redownload the updates thru the colleges T1 line.

I have no control over the college network, but am thinking there is
some problem, but don't know how to identify exactly where it is.
Why would 18 of 20 machines work fine. The physical network has
been setup by the College MIS as one network with three class C's.

202.128.71.0, 202.128.72.0, and 202.128.73.0

The 71 and 73 networks are dhcp and provided to the classrooms,
generally, you will get about an even mix of 71 and 73 numbers in
the same room, and the only link between the networks is the router
that connects to the colleges T1 (only a 10MB router), so any
transfers would drop to 10MB when between machines on different
class C's. (I've added special routing entries in my labs to avoid this)
But seeing that this is happening between machine on the same
class C.

Today's problem is with 202.128.73.253 and 202.128.73.211. They
get the !<10> error, but both transfer fine to 202.128.72.68.

+----------------------------------------------------------+
  Michael D. Setzer II - Computer Science Instructor
  Guam Community College Computer Center
  mailto:mikes@kuentos.guam.net
  http://www.guam.net/home/mikes
  Guam - Where America's Day Begins
+----------------------------------------------------------+

http://setiathome.berkeley.edu
Number of Seti Units Returned: 9,931
Processing time: 21 years, 294 days, 15 hours, 18 minutes
(Total Hours: 191,031)

-- 
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list

• Previous message: Phil Schaffner: "Re: Need Octave Forge rpm"
• In reply to: Phil Schaffner: "Re: special install situation"
• Next in thread: Alexander Dalloz: "Re: traceroute error !<10>"
• Reply: Alexander Dalloz: "Re: traceroute error !<10>"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]