- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-09-2011 04:41 AM - edited 03-11-2019 02:22 PM
Here's the problem we've encountered, was hoping you guys could help me debug what the heck is going on:
In the below diagram, computers on 10.20.1.x cannot ping anything on the 10.20.3.x subnet (Including Firewall B).
However, Firewall A can ping Firewall B and the .3.x server so it's not a VLAN routing problem.
The strange thing is that according to the ASDM Packet trace on Firewall A (Screenshot), packets are allowed to travel freely between the internal pc's and the .3.x subnet servers.
I'm positive the fault lies with Firewall A, but the running config file is way too big to post here so I'll try dig out what looks relevant.
We have the following NAT rule which I would have though was the solution but that doesn't seem to help:
static (inside,bt_dmz) 10.20.1.0 10.20.1.0 netmask 255.255.255.0
(bt_dmz is the interface on Firewall A that runs through to the 10.20.3.x subnet)
We also have icmp permitted:
icmp permit 10.20.1.0 255.255.255.0 inside
icmp permit 10.20.2.0 255.255.255.0 dmz
icmp permit 10.20.3.0 255.255.255.0 bt_dmz
Access list rule:
access-list outside_access_in_dmz extended permit icmp any any
If Firewall A can ping Firewall B and the server, it feels like routing between VLANS is ok but why then can I not ping the server on that VLAN from my internal PC? Could there be a conflicting rule?
Solved! Go to Solution.
- Labels:
-
NGFW Firewalls
Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:28 AM
Could you please add this:
access-list vlan1_bt_dmz_access_in extended permit icmp any any
this should resolve the icmp issue and check packet-tracer again.
Thanks,
Varun
Varun Rao

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-09-2011 04:50 AM
Hi Wez,
I was answering to your thread but I guess you moved it by then
Well I woudl say take captures and analyse, thats the best thing. Check where the packets are getting dropped. Also you woudl need a nat statement for translating the source traffic On firewall 1, something like this:
nat (inside) 10 10.20.1.0 255.255.255.0
global (dmz) 10 interface
On the firewall 2, you shoudl have the following:
static (inside,dmz) 10.20.1.0 10.20.1.0
I would request you to kindly try it and take captures. On firewall 2 as well you need the access-list going from dmz to inside, so plz make sure.
For captures:
https://supportforums.cisco.com/docs/DOC-17814
Hope this helps.
Thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-09-2011 05:58 AM
Hey, sorry about that. I posted in routing but fixed an issue soon after which made it seem more like a firewall issue than a routing one so I recreated the post here.
With Firewall B, I'm not sure it needs the change you recommended as it's only role is to serve traffic between .3.x servers and the internet. It doesn't actually fall between any internal traffic. Though I might be wrong on that one.
With Firewall A, we have the current rules:
static (inside,bt_dmz) 10.20.1.0 10.20.1.0 netmask 255.255.255.0
access-list inside_nat0_outbound extended permit ip any 10.20.1.16 255.255.255.240
nat (inside) 0 access-list inside_nat0_outbound
nat (inside) 101 0.0.0.0 0.0.0.0
global (outside) 101 interface
global (BT_Outside) 103 interface (Redundant)
global (bt_dmz) 102 interface
So if I rejig your rules, should I still add the following? :
nat (inside) 101 10.20.1.0 255.255.255.0
And does this make the global rule ( global (bt_dmz) 10 interface ) redundant?

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-09-2011 06:04 AM
Hi Wez,
Yeah I had a look at it again, and i guess if the traffic for the server is not going through the 2nd firewall, then you don't need any config on it. From the config that you have , i can only see one thing that needs to be added:
global (bt_dmz) 101 interface
You can try it, if it still does not work, take captures and trace the packets.
Thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-09-2011 07:08 AM
Here are the results from the packet capture wizard. It looks like packets are going one way but not coming back the other.
I'm guessing that means there are access rules allowing traffic from my pc to the server, but not in the other direction?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 02:15 AM
What does global (bt_dmz) 101 interface actually do? As far as I understood, it assigns one of the physical ports on the box as an interface.
The outside interface assigned to 101 is already used to connect out to the internet. 102 is where the bt_dmz connects up to.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 03:28 AM
Hi Wez,
The statement is for patting the inside users to the ip address of the bt_dmz interface:
nat (inside) 101 0.0.0.0 0.0.0.0
global (bt_dmz) 101 interface
It is the nat ID's that needs to be matched, 101 is the nat id for the global statement corresponding to the one in the nat statement.
As you can see in the captures, when you ping, the echo request is going fine through the firewall, but we do not see and replies coming back for it, I would suggest you use the global statetment suggested and then take the captures again.
Let me know i you have any issues.
Thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 04:13 AM
Added the rule and re-pinged, still have the same problem. The request is going one way, but not the other.
New globals/nats look like this:
global (outside) 101 interface global (BT_Outside) 103 interface global (bt_dmz) 101 interface nat (inside) 0 access-list inside_nat0_outbound nat (inside) 101 0.0.0.0 0.0.0.0
Ran the packet tracer tool on the asa, From inside to bt_dmz, it works as expected but the other way I get an access list failure. (Though I'm not sure it is because I have the wrong interface selected) - Screen shot
The access list it drops on is the default any - any - deny rule in the inside interface.
Could it be something like a static route that had been set up ?
We had once tried to get the BT_dmz to go out via BT_outside interface, but abandoned that plan. I don't think there are any rules left that would cause issue though.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 04:38 AM
Can you provide me the access-list that you have applied on the inside interface??
Thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:08 AM
Sure,
access-list outside_access_in_dmz extended permit icmp any any
access-list outside_access_in_dmz extended permit tcp any object-group EmailServers eq smtp
access-list outside_access_in_dmz extended permit object-group WebsiteTraffic any object-group WebServers
access-list outside_access_in_dmz extended permit udp any host 10.20.1.5 eq sip
access-list Internal standard permit 192.168.2.0 255.255.255.0
access-list Internal standard permit 10.20.1.0 255.255.255.0
access-list Internal standard permit 10.20.2.0 255.255.255.0
access-list Internal standard permit 10.1.1.0 255.255.255.0
access-list Internal standard permit 10.1.10.0 255.255.255.0
access-list inside_access_in extended permit ip any any
access-list dmz_to_inside extended permit ip any any
access-list inside_nat0_outbound extended permit ip any 10.20.1.16 255.255.255.240
access-list vlan1_bt_dmz_access_in extended permit ip any any
access-list inside_access_in_1 extended permit ip any 10.20.3.0 255.255.255.0
access-list inside_access_in_1 extended permit ip any any
access-list BT_Outside_access_in extended permit icmp any any
Nothing is standing out as wrong from what I can see

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:12 AM
Hi Wez,
Could you also send:
show run access-group as well.
thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:20 AM
Sorry, missed those.
access-group outside_access_in_dmz in interface outside
access-group inside_access_in_1 in interface inside
access-group BT_Outside_access_in in interface BT_Outside
access-group dmz_to_inside in interface dmz
access-group vlan1_bt_dmz_access_in in interface bt_dmz
I have a big old list of service and network objects if you want them too, though there's nothing really in there relating to either the .3.x subnet or the bt_dmz

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:28 AM
Could you please add this:
access-list vlan1_bt_dmz_access_in extended permit icmp any any
this should resolve the icmp issue and check packet-tracer again.
Thanks,
Varun
Varun Rao
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:49 AM
Added that in, and re-ran the packet capture (Results) but with little success.
Would it help if I private messaged you the full running config?

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-12-2011 06:51 AM
Sure Wez, go ahead. You can PM me the config.
-Varun
Varun Rao
