Results 1 to 2 of 2
Hi! The Scenario: A Blade running VMWare with several Virtual Machines. Some of them running Windows 2003 Server and others running Red Hat 5.4. The Blade is directy connected to ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
- 08-26-2010 #1
- Join Date
- Aug 2010
[Tcptraceroute/Traceroute/Ping] Windows VMs & Linux VMs
A Blade running VMWare with several Virtual Machines. Some of them running Windows 2003 Server and others running Red Hat 5.4.
The Blade is directy connected to a Cisco Switch that has several Cisco Switchs connected to it. To this point all was running perfectly fine.
A new Cisco Switch is purchased, configured and then the Blade and all the other Cisco Switchs are connected directly to this New Cisco Switch.
Something really strange begins to happen:
* PING fails from Linux Virtual Machines to all the Switchs except the New Cisco Switch.
* Traceroute/Tracert from Linux Virtual Machines to the New Cisco Switch never reaches the destination.
* PING succeded from Windows Virtual Machines to all the Switchs except the New Cisco Switch. Seems important to note that the FIRST PING made, fails.
* Tracert from Windows Virtual Machines to the New Cisco Switch reaches the destination in 2 Hops.
I start to investigate a little with Google and i found the Linux command "tcptraceroute".
The first time i use it to reach any of the old Routers i get something like this:
# tcptraceroute UUU.XXX.YYY.ZZZ traceroute to UUU.XXX.YYY.ZZZ (UUU.XXX.YYY.ZZZ), 30 hops max, 40 byte packets 1 (AAA.BBB.CCC.DDD) 1.399 ms 0.488 ms * 2 * * * 3 * * * 4 * * * 5 * * * 6 * * * 7 * * * 8 * * * 9 * * * 10 * * * 11 * * * 12 * * * 13 * * * 14 * * * 15 * * * 16 * * * 17 * * * 18 * * * 19 * * * 20 * * * 21 * * * 22 * * * 23 * * * 24 * * * 25 * * * 26 * * * 27 * * (UUU.XXX.YYY.ZZZ) 2005.795 ms
After doing that, making PING, Traceroute and Tracert from the Linux Virtual Machines to the same IP works fine and without errors.
I've run tcptraceroute from the Linux Virtual Machines to all the Switchs to make them work and now its all working fine!
However i cant understand two things:
1) Why the Linux Virtual Machines couldnt be able to PING or Traceroute to the old Switchs before running the tcptraceroute?
2) Why the tcptraceroute made 27 HOPs to reach a destination that should take only 2 HOPs?
Please any help would be really appreciated.
Thanks in advance,
- 08-27-2010 #2
- Join Date
- Apr 2009
- I can be found either 40 miles west of Chicago, or in a galaxy far, far away.
RCA - Root Cause Analysis. Inclusion of new switch and things go fubar. What happens when the new switch is removed (de-integrated) from the system? I'd say that it's about 100% certain that something in the configuration of the new switch has caused this problem. Start there.Sometimes, real fast is almost as good as real time.
Just remember, Semper Gumbi - always be flexible!