cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
546
Views
1
Helpful
7
Replies

DNS only works on one host?

Seanmorrow
Level 1
Level 1

So, I had DNS working, and I started putting FQDNs in ACLs. All were working and populating "show dns". Then I rebooted the firewall a few weeks later and now it's not resolving most hosts - saying they aren't active. The below code is partial and I can't post the whole, but don't be afraid to ask for any other non-compromising parts. There are plenty of others that don't work. This is just my best example since they share a group - yet only one is active.

For example:

object network fqdn_alertmedia_com_dashboard
    fqdn v4 dashboard.alertmedia.com
object network fqdn_sqs.us-east-1.amazonaws.com
    fqdn v4 sqs.us-east-1.amazonaws.com
object-group network AlertMedia_Servers
    network-object object fqdn_alertmedia_com_dashboard
    network-object object fqdn_alertmedia_aws
(object group is in an ACL and applied)

ASA-5516# sh dns
 Name: dashboard.alertmedia.com
  Address: 3.214.247.59 TTL 00:01:02
  Address: 18.205.85.59 TTL 00:01:02
(this *all* of the output)

(It doesn't respond to pings, but you can see it is capable of resolving)
ASA-5516# ping sqs.us-east-1.amazonaws.com
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 3.239.232.46, timeout is 2 seconds:
?????
Success rate is 0 percent (0/5)

ASA-5516# ping google.com
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 142.250.217.142, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/10 ms

(I even added google.com to the object group and "show dns" still won't show it.)

1 Accepted Solution

Accepted Solutions

asa# sh access-list <fqdn acl name>

did you see other ACL FQDN as inactive ? 
MHM

View solution in original post

7 Replies 7

can you ping DNS server from the mgmt interface or from interface use connect to host that filter by ACL?
share 
show run dns
MHM

DNS "works" and resolves addresses, just not for any object except the one. There was an internal server, but I removed it for troubleshooting purposes.

ASA-5516# show run dns
dns domain-lookup Inside
dns domain-lookup Outside
dns domain-lookup Outside2
DNS server-group DefaultDNS
name-server 8.8.8.8 Outside
ASA-5516# ping community.cisco.com
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 18.154.144.23, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/4/10 ms
ASA-5516# ping reddit.com
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 151.101.1.140, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/6/10 ms
ASA-5516# ping 8.8.8.8
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 8.8.8.8, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/10 ms

 

and the user use 8.8.8.8 as DNS server ?
MHM

They are not, but as said: I removed the internal (user) DNS to make sure it wasn't an issue with our DNS. I just haven't put it back yet since I feel 8.8.8.8 is a more reliable test. Once I get DNS working with the other network objects I'll put it back.

asa# sh access-list <fqdn acl name>

did you see other ACL FQDN as inactive ? 
MHM

I think you just hit the problem. They're applied to an ACL, but the service that uses those ACLs was *not* running. The one that was resolving was doing so because of a different ACL for a different purpose - that just also happened to be here.

Oof on me. Thanks for the assist.

friend you are so welcome 
MHM

Review Cisco Networking for a $25 gift card