NAT OVERLOAD
I once had a small office setup where users suddenly lost internet access. The router was up, local traffic worked fine, but anything beyond the LAN was dead. At first glance, it looked like the ISP was down, but digging deeper revealed the real culprit: a misconfigured NAT overload (PAT).
🔍 How I Diagnosed It
I started with the basics:
show ip interface brief→ confirmed all interfaces were up.ping 8.8.8.8from the router worked, but not from internal hosts.show ip nat translations→ came back empty.
That was the red flag. NAT wasn’t translating internal IPs to the public IP. Without translations, users couldn’t reach the internet.
✅ How I Fixed It
Defined the ACL for internal traffic:
Codeaccess-list 1 permit 192.168.1.0 0.0.0.255Configured NAT overload:
Codeip nat inside source list 1 interface GigabitEthernet0/0 overloadSet interface roles:
Codeinterface GigabitEthernet0/1 ip nat inside interface GigabitEthernet0/0 ip nat outsideVerified NAT translations:
Codeshow ip nat translations
Once this was in place, NAT entries started populating, and users could access the internet again.
📌 Why This Is Important
This issue looks simple but happens a lot—especially when configs are copied from one router to another or after a reset. NAT isn’t plug-and-play; it needs precise alignment between ACLs, inside/outside interface roles, and overload rules.
For me, the lesson was: always check NAT translations early when internet access breaks. If the router itself can ping out but users can’t, it’s almost always NAT misconfiguration. Fixing it quickly restores productivity and avoids unnecessary ISP blame.
No comments:
Post a Comment