Procurve 5412 zl switch configuration issue

[andy_johnson(cnews)]

We currently have two procurve 5412 zl devices that are acting as core switches for our network. They are using rip v2 as their routing protocol.
The problem I am facing is that I am unable to send traffic to one of our networks "192.168.14.0/24" unless the traffic is tagged as a member of vlan 100.
I can see a dynamically added entry for the above network in the routing table of both core devices and the route is associated with vlan 100. What I need is access to the 192.168.14.0 network regardless of vlan.
If the information I have provided does not provide enough detail I am happy to send configs across etc.

Thanks in advance!!!!

Andy Johnson

P.S. This thread has been moved from Switches, Hubs, Modems (Legacy ITRC forum) to ProCurve / ProVision-Based. - Hp Forum Moderator

[Pieter 't Hart · Accepted solution]

Hi Andy,

normally vlans and subnets are mapped 1-1 like :
vlan-10 192.168.10.0/24
vlan-14 192.168.14.0/24
a router (dedicated router or routing switch) must pass trafic from one subnet to the other.

>>> What I need is access to the 192.168.14.0 network regardless of vlan <<<

looks like data is sent to the wrong router.
or the routing config is not complete.

>>> The problem I am facing is that I am unable to send traffic to one of our networks "192.168.14.0/24" unless the traffic is tagged as a member of vlan 100. <<<
one of the switches should have an interface in vlan-100 and do the routing to this subnet.
the sending host should send data to the router in its own vlan.

yes more detailed switch(es) config will help.

[andy_johnson(cnews)]

Hi Pieter,

I think you may me right about the routing config not being complete. unfortunately I unable to see where?
I have attached a txt file including running-configuration and routing tables for your perusal I hope that this will provide a bit more information for you. If I can provide any more detail please let me know and I will be happy to help.

Many Thanks

Andy Johnson

[Pieter 't Hart]

maybe this explains what's wrong?
>>>
vlan 100
name "VLAN100"
untagged E2,E5-E11,E13-E16,E18,E20-E23,F2-F16,G1-G24,H1-H24
ip helper-address 10.0.20.1
ip helper-address 10.0.20.2
ip address 10.0.2.3 255.255.224.0
<<<
the interface has a subnetmask of 255.255.224.0
all ip-adresses 10.0.0.1 to 10.0.15.254 will be considdered "local" to this interface, and packets will not be sent through a gateway to other vlans, but directly to the endpoint.
The "rip-learned" route using device 10.0.0.1 is ignored (with rip all routers must use the same mask for a specific subnet; here 255.255.255.0).

[andy_johnson(cnews)]

Hi Peiter,

I think I understand what you are saying but can I double check.

The rip route is being ignored because the subnet mask /19 associated with vlan 100 is incorrect.

Do you think it would be prudent to change the subnet mask? /17 perhaps?

I have attached a traceroute from our core switch and from a non vlan 100 machine to a device on the 192.168.14.0 network. does this support your theory?

Many Thanks

Andy Johnson

[Pieter 't Hart]

no it doesn't, but it show some other interesting info.

from the core switch the traceroute DOES adress the router 10.0.0.1, but this sents the data via (public internet?) routers/firewalls 64.214.251.241 and 64.214.35.142 to the destination 192.168.14.7.
whereas the non-vlan100 client does the same, but gets no response.

looks like this is a remote location?

then possibly the remote location can find the route back to 10.0.0.0/19
but cannot to 10.0.35.3 (10.0.35.0/24 or 10.0.32.0/19 )

[andy_johnson(cnews)]

Hi Pieter,

Thanks for your help with this however I am still at a loss regarding this issue. The traceroute from the non vlan 100 machine proves that traffic not tagged as vlan 100 is dropping off the network somewhere after it comes off our mpls (64. addresses).
This supports your theory of no route back for non vlan 100 traffic. I guess what I need to do is get some configuration details for the (192.168.14.0)connectivity devices and go from there.
Is it ok to update you once I have that information?

Thanks Again

Andy Johnson