Removing VLAN - and routing to Node

01070265:3: The VLAN (/Common/VLAN_xx) cannot be deleted because it is in use by a self IP (xx.xx.xx.xx).

have you tried to remove selfip and then vlan?

if not working, may you try to modify configuration file and force to reload configuration?

sol13030: Forcing the mcpd process to reload configuration

01070265:3: The VLAN (/Common/VLAN_xx) cannot be deleted because it is in use by a self IP (xx.xx.xx.xx).

have you tried to remove selfip and then vlan?

if not working, may you try to modify configuration file and force to reload configuration?

sol13030: Forcing the mcpd process to reload configuration

But have you tried to remove the self-IP and then the VLAN, as Nitass recommended?

As long as you have objects that reference a specifically defined subnet, you're not going to be able to delete the self-IP and consequently the VLAN. As you've found, however, you don't need to delete the VLAN and self-IP to get to these routed nodes. You don't even really need a default route as long as you have A route. So what you're probably going to have to do is (perhaps slowly) migrate existing pools to the routed members, and once all of the local stuff is gone, then you can delete the self-IP.

Hi Kevin,

How is it going to use the routing table, when it thinks it is a directly connective network.

I have never had this situation before, so I dont know - It just seems like it wont, because a directly connected network might have priority over a route

Graham

You have pool members that are in a different subnet and you have a route to that subnet through either a default or other specific route. If you create a route to another subnet then you should have a VLAN and self-IP in the local subnet (the subnet that exists between the F5 and the router) - so the device knows which physical path to take to get to that network. You could alternately send a route through a specific VLAN, but it certainly doesn't hurt to have a local self-IP.

i did a bit test (below) but it did not work. so, it seems you have to modify configuration file directly (i.e. bigip_base.conf and bigip.conf) and then force mcpd to reload the configuration.

 existing pool

root@ve10(Active)(tmos) list ltm pool foo
ltm pool foo {
    members {
        200.200.200.101:http { }
    }
}

 existing connected subnet

root@ve10(Active)(tmos) list net self 200.200.200.16/24
net self 200.200.200.16/24 {
    allow-service default
    vlan internal
}

 new vlan

root@ve10(Active)(tmos) create net vlan newvlan interfaces add { 1.2 { tagged } }

 new selfip

root@ve10(Active)(tmos) create net self 222.222.222.16/24 vlan newvlan allow-service default

 new route but error

root@ve10(Active)(tmos) create net route 200.200.200.0/24 gw 222.222.222.254
01070666:3: Static route duplicates Self IP 200.200.200.0 / 255.255.255.0 implied route

 delete selfip but error

root@ve10(Active)(tmos) delete net self 200.200.200.16/24
01070605:3: Cannot delete IP 200.200.200.16 because it would leave a pool member (pool foo) unreachable.

i did a bit test (below) but it did not work. so, it seems you have to modify configuration file directly (i.e. bigip_base.conf and bigip.conf) and then force mcpd to reload the configuration.

 existing pool

root@ve10(Active)(tmos) list ltm pool foo
ltm pool foo {
    members {
        200.200.200.101:http { }
    }
}

 existing connected subnet

root@ve10(Active)(tmos) list net self 200.200.200.16/24
net self 200.200.200.16/24 {
    allow-service default
    vlan internal
}

 new vlan

root@ve10(Active)(tmos) create net vlan newvlan interfaces add { 1.2 { tagged } }

 new selfip

root@ve10(Active)(tmos) create net self 222.222.222.16/24 vlan newvlan allow-service default

 new route but error

root@ve10(Active)(tmos) create net route 200.200.200.0/24 gw 222.222.222.254
01070666:3: Static route duplicates Self IP 200.200.200.0 / 255.255.255.0 implied route

 delete selfip but error

root@ve10(Active)(tmos) delete net self 200.200.200.16/24
01070605:3: Cannot delete IP 200.200.200.16 because it would leave a pool member (pool foo) unreachable.

Change the Subnet Mask on the Self-IP from a /24 to a /32 (255.255.255.255). This will then allow you to delete the Self-IP from the configuration

thanks, it's nice.

 existing configuration

[root@B3600-R66-S40:Active] config  b vlan internal list
vlan internal {
   tag 423
   interfaces tagged 1.1
}
[root@B3600-R66-S40:Active] config  b self 200.200.200.40 list
self 200.200.200.40 {
   netmask 255.255.255.0
   vlan internal
   allow default
}
[root@B3600-R66-S40:Active] config  b pool foo list
pool foo {
   members 200.200.200.101:http {}
}

 deleting

[root@B3600-R66-S40:Active] config  b self 200.200.200.40 netmask 255.255.255.255
[root@B3600-R66-S40:Active] config  b self 200.200.200.40 delete
[root@B3600-R66-S40:Active] config  b self 200.200.200.40 list
BIGpipe self IP address query error:
   01020036:3: The requested self IP address (200.200.200.40) was not found.
[root@B3600-R66-S40:Active] config  b pool foo list
pool foo {
   members 200.200.200.101:http {}
}

This worked from me to. Thank you all! Just to re-highlight for anyone who sees this page in the future.

1. Delete the floating ip(s)
2. Change non-floating ips to /32 (255.255.255.255 subnet mask) ( on all devices in the cluster )
3. Delete non-floating ips

Now the F5 will attempt route to the nodes in question that were causing the conflict.