cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
907
Views
0
Helpful
2
Replies

MS NLB ping keep interrupt

hisham.khreisat
Level 1
Level 1

Hello there,

Please I need your help in the following setup:

I have two TMG Server connected to switch 3750, The problem is when we created NLB for the servers(Multicast), After little time, the ping keep interrupt when we ping the default gateway for the servers.

The intteruption happened with

general failure

general failure

general failure

request timed out

request timed out

request timed out

we configured the static MAC Address  and arp on the switch, but the interruption still exit.

This problem appears when we put the gateway on NIC card configuration, but if we removed it or make route to point to gateway, the problem solved.

have anyone any simliar issue or any solution for it.

Why the NIC Card keep diconneted like this???!!! Is it a Network issue or from TMG Servers ??? Is this problem from the NIC Card ???

How I can isolate the problem? or troubleshoot it ??

Please advice...

2 Replies 2

IAN WHITMORE
Level 4
Level 4

Did you see the failures...

general failure

general failure

general failure

request timed out

request timed out

request timed out

...on the server on the switch or on your PC?. If they were on the server, what happens if you start a ping -t to the gateway. Do you failures stop?

Could be an ARP timing issue. NLB on Microsoft is not the best and just floods the network, so when the MAC's timeout on the switches, sometimes they lose connectivity if there is no traffic on the servers. If you put a ping -t from the server to the gateway, it keeps the MAC active. This is just a work-around to pinpoint the problem. If it is this, there are things you can do like increase the MAC timeout on the switches. However this can only be applied to 6500 switches I think.

Of course the static MAC address might not work because NLB uses a different MAC. Here is a link on putting static addresses for NLB:

http://www.cisco.com/en/US/products/hw/switches/ps708/products_configuration_example09186a0080a07203.shtml

Also, in the switch logs do you get any errors? Are the interfaces on the switch/server hard coded for speed/duplex? Is the switch IOS up-to-date? Are the server NIC drivers up-to-date?

There could be many things...

HTH,

Ian

At first I would like to say something here,

I did a continues ping from the server to the gateway, one of the Nodes its ping keep interrupt.Yes I got General Failure from Server side.

Switch and Servers have latest version.

the logs on the switch just give me the following:

May 19 21:33:48.120: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to up

May 19 21:36:12.136: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to down

May 19 21:36:14.141: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to up

May 19 21:36:56.134: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to down

May 19 21:36:58.139: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to up

May 19 21:38:35.455: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to down

May 19 21:38:37.460: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to up

May 19 21:39:50.156: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to down

May 19 21:39:52.160: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to up

May 19 21:41:14.797: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/12, changed state to down

I didn't get anything else, the server and switch work on auto speed and auto duplex, and worked on FULL duplex / 100 Mbps Speed.