cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
7713
Views
5
Helpful
6
Replies

problem pinging cluster ip of MS NLB in multicast mode on vmware

Jody Sudbury
Level 1
Level 1

I've been trying to get NLB running in my environment. My servers are running server 2008 in VMware. I have a pair of catalyst 3560X switches running HSRP and routing traffic.

I understand the problem with ARPs and mixing a unicast IP with a multicast mac. I've followed these instructions from teh cisco site:

http://www.cisco.com/en/US/products/hw/ ... 7203.shtml

After i did that stuff i was able to ping the cluster ip from a different subnet (couldn't before). Once I shut down one of the machines in the NLB cluster, I was no longer able to ping the cluster IP (both VMs in cluster were showing converged) from outside the subnet. I then tried pinging from the switches directly and found that I could my NLB cluster IP (10.1.0.60) from one switch (core2), but not from the other(core1), even when i source my ping from vlan 100 (the same the cluster is in).

I ran a debug packet and debug icmp and there is no output at all, which i find strange.

The two switches each have a connection to 2 ESXi hosts and the port #s are the same for both.

03bf.0a01.003c is the cluster MAC and 10.1.0.60 the cluster IP. The servers are in vlan 100.

Relevant config is below:

Switch 1 (not working)


arp 10.1.0.60 03bf.0a01.003c ARPA
mac address-table static 03bf.0a01.003c vlan 100 interface GigabitEthernet0/15 GigabitEthernet0/16

Switch 2 (working)


arp 10.1.0.60 03bf.0a01.003c ARPA
mac address-table static 03bf.0a01.003c vlan 100 interface GigabitEthernet0/15 GigabitEthernet0/16

Show command output

The show ip arp and show mac address-table output are the same on each switch:

core2#show ip arp 10.1.0.60
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  10.1.0.60               -   03bf.0a01.003c  ARPA

core2#show mac address-table address 03bf.0a01.003c
          Mac Address Table
-------------------------------------------

Vlan    Mac Address       Type        Ports
----    -----------       --------    -----
100    03bf.0a01.003c    STATIC      Gi0/15 Gi0/16
Total Mac Addresses for this criterion: 1

debug output when I ping cluster ip 10.1.0.60

On Switch 1 (no output):


core1#debug ip packet 95
core1#debug ip icmp
core1#ping 10.1.0.60
pak 44FE900 consumed in input feature , packet consumed, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0
pak 44F8054 consumed in input feature , packet consumed, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0

On switch2 (works):


core2#ping 10.1.0.60
Jul 19 09:38:34.702 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, input feature, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0
Jul 19 09:38:34.702 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, rcvd 2
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, stop process pak for forus packet
Jul 19 09:38:34.710 EDT: ICMP: echo reply rcvd, src 10.1.0.60, dst 10.1.0.3
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, input feature, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, rcvd 2
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, stop process pak for forus packet
Jul 19 09:38:34.710 EDT: ICMP: echo reply rcvd, src 10.1.0.60, dst 10.1.0.3
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, input feature, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, rcvd 2
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, stop process pak for forus packet
Jul 19 09:38:34.710 EDT: ICMP: echo reply rcvd, src 10.1.0.60, dst 10.1.0.3
Jul 19 09:38:34.710 EDT: IP: s=10.1.0.60 (Vlan100), d=10.1.0.3, len 100, input feature, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/4/9 ms

pak 3EAD104 consumed in input feature , packet consumed, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0
pak 44F3BC0 consumed in input feature , packet consumed, MCI Check(63), rtype 0, forus FALSE, sendself FALSE, mtu 0

Does anyone have any idea of what i'm missing here? Hardware is identical. IOS is the same. Configs are the same. I'm not sure what else to look at...

Thx

6 Replies 6

didkovsky
Level 1
Level 1

how about show spanning-tree summary?

Same issue accured at my end and still did not provide solution.

please share solution.

Hi All,

 

What mode do you have the MS NLB cluster running in? Unicast or Multicast?

 

 

MS NLB running in IGMP

thanks cisco support for this blog issue is resolved.
Review Cisco Networking for a $25 gift card