cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
919
Views
0
Helpful
3
Replies

Proble Fail after switchover

Hi All,

Im just going through some failover testing with a pair of ACE  4710 appliances configured as a fault tolerant pair.

After I just disconnect the ethernet cables from the active ACE, all VIP`s are transfered properly to standby ACE, but when I check the real server status, all servers are not Inservice, this is showing Probe Failed. If I select all real servers, and click in Activate, all servers becomes working fine.

What`s is wrong on my config and how to fix that urgently?

ft interface vlan 543

  ip address 192.168.100.1 255.255.255.252

  peer ip address 192.168.100.2 255.255.255.252

  no shutdown

ft peer 1

  heartbeat interval 300

  heartbeat count 10

  ft-interface vlan 543

  query-interface vlan 542

ft group 1

  peer 1

  peer priority 200

  associate-context Admin

  inservice

ft track interface TRACK_VLAN542

  track-interface vlan 542

  peer track-interface vlan 542

  priority 2

  peer priority 2

interface port-channel 1

  ft-port vlan 543

  switchport trunk native vlan 542

  switchport trunk allowed vlan 542

  port-channel load-balance src-dst-port

  no shutdown

3 Replies 3

vpurbiya
Level 1
Level 1

Hi Jeff

Hope you are doing good,

So once the failover happens all the vip are moved to the backup, did you verify they are in service ?

The situation that you are talking about the servers going in probe failed mode, is it after failover or before failover.

So far i could think of an connectivity issue between the secondary node and the server vlan, the arp msg are somehow not making it to the servers. or either not generated at all

Reason why it works after you forcefully make the servers active, it actually forces an ARP request sent out on the lan,

So can you verify via taking capture on the server at the time you are doing failover , that you get an ARP msg for the same or not.

Ok Vikas, just to understand, when I can configure the backup servers?

The problem happens after failover.

In the trunk VLAN, I need to allow the corporate VLAN and FT VLAN? Two machines are able to reach via ICMP. What need to be allowed to allow that?

Ok

So these are diff things

1. backup server

it either can be at rserver level or serverfarm level. The sole intention is to give a redundancy to real server or entire serverfarm just in case the primary goes down.

http://www.cisco.com/en/US/docs/interfaces_modules/services_modules/ace/v3.00_A1/configuration/slb/guide/rsfarms.html#wp1000428

2. Ft is just the communication between the ACE, FT packet never needs to go beyond ACE.

So if you are talking about the trunk interface which is connecting the ACE, yes you need to allow both FT and corporate vlans on that.

If you are talking about the downstream layer 2 connectivity beyond ACE, you dont need FT vlan there

3. Now regarding your original query:-

After failover, can you check the show arp on the secondary (now active node), whether it has the arp entry or not.

Also check for the show probe detail and check why it is showing failed. , it should give you the error code

Review Cisco Networking for a $25 gift card