What VLAN does ping use by default

[CiscoBrownBelt]

Does ping use VLAN 1 by default?

So basically I have Site A with let's say a edge router and 4 switches behind it, that is connected to another site B just the same. All switches at Site A can ping let's say 10.10.10.5 address at Site B except for one switch. That switches and addresses at Site B except that one address. The switch is trunked just the same, has all the same VLANs although it should not matter since traffic is going across WAN, and using same default-gateway just as the other 3 switches at the site that are able to hit everything. 

The ACLs are not blocking the traffic.

Does anyone know anything else that may come to mind that I should look at?

[chrihussey · Accepted solution]

Just some questions:

1- Is the 10.10.10.5 IP the only address at site B the switch cannot ping and it can ping others?

2- Can the switch ping anything else on the 10.10.10.5's subnet?

3- Is the switch that can't ping the IP on the same VLAN as the other switches?

4- What are the results of a trace route from the 10.10.10.5 switch? It should tell you where it is being stopped.

Finally, If you have everything set the same switchwise at site A, then there is the possibility the ping is getting to site B, it may just be the response is not getting back.

Hope this helps.

[CiscoBrownBelt]

Ok well looks like when doing pings, the ping packet will have  a source address using the subnet for SVI 4 (10.10.4.X) and a ACL was blocking it.

The thing I would like to understand is what makes the switch use that source address to ping? I believe I could leave the ACL with the permit 10.10.X statement pertaining to the int g0/0.4 on the edge router at the site, or remove all the SVIs except management SVI10 from the switch (or all switches) as all the switches point to the edge router, and the edge router does the routing with ROAS. Would this be the best option?

Just to answer your questions prior to me allowing that subnet to the ACL:

1- Is the 10.10.10.5 IP the only address at site B the switch cannot ping and it can ping others?

Yes

2- Can the switch ping anything else on the 10.10.10.5's subnet? No

3- Is the switch that can't ping the IP on the same VLAN as the other switches?

Yes all switches have SVIs and vlans for vlans 4,6,8, 

4- What are the results of a trace route from the 10.10.10.5 switch? It should tell you where it is being stopped. stops at .1 on the Edge router of the site.

Finally, If you have everything set the same switchwise at site A, then there is the possibility the ping is getting to site B, it may just be the response is not getting back.

Hope this helps.

[chrihussey · Accepted solution]

As a general rule a switch or router will use the IP address of the interface it uses to reach the destination. I understand that you have a management IP and a default gateway specified on the switch, but it is using the SVI4 interface. So something in its routing is telling it to use that interface. Perhaps with more information or if you post its config we can figure that part out.

Anyhow, if you want it to use the management IP for the ping you can specify the source address by doing an extended ping.

Regards

[CiscoBrownBelt]

Yea I don't have the config to paste on here.

Yes source ping is what I was doing to test and noticed what was going on even before looking at the logs.

Yes I don't know why it would use that SVI interface as default-gateway is using the management ip address. The configs of all the switches and routers are not too too complicated especially the switches (few SVIs, either default-gw or 0.0.0.0 route to edget Rt .1 mgmt IP addrress, no ACLs at switch only Rt, etc.).

I will revisit this issue later next week.