Hello,

We have a problem. We have a router which has, amongst others, ports 80 and 443 open. Both are pointing to one machine running apache2.

However, https on 443 works, but http on 80 doesn't.

nmap shows port 80 as filtered (why?), and port 443 as open.

I have stopped the firewall on the web server. I have made no changes to the apache2 config since installing it. Still no "It works" page!

Any help gratefully received!

Kind regards

Dani AI

Generated

Short expert note for future readers:

’s case—port 443 reachable but nmap showing port 80 as “filtered” until a router reboot—is a classic symptom where packets for HTTP never get a response from the network edge. “Filtered” means nmap saw no reply (not an RST), so the packet was likely dropped by the ISP or the router rather than refused by the web server. Possible causes include ISP port blocking/transparent proxy, the router’s remote‑management or proxy using port 80, a stale NAT/conntrack state on the router, or an internal router firewall rule.

Practical checklist (run on the server first, as suggested):

# confirm apache is listening on 80
ss -ltnp | grep :80

# test locally and via LAN IP
curl -I http://127.0.0.1/
curl -I http://SERVER_LAN_IP/

# check kernel/firewall rules
sudo iptables -L -n -v
sudo ufw status

If those prove the server is serving OK, verify whether traffic arrives from the WAN. On the webserver run:

sudo tcpdump -n -i any tcp port 80

Then, from an external network (mobile data or a remote shell) try curl -I http://YOUR_PUBLIC_IP/ or telnet YOUR_PUBLIC_IP 80. If tcpdump shows nothing, the router or ISP is dropping/never forwarding the packets. If packets arrive but there’s no server response, recheck firewall and virtualhost/listen settings.

Practical mitigations and next steps: disable any router remote‑management on port 80, change router admin to a nonstandard port, update router firmware, and capture tcpdump + router logs before contacting the ISP (evidence makes escalation faster). If reboot fixed it once, suspect stale NAT/conntrack or a router bug; collect the packet captures so the provider can diagnose persistent recurrence.

Recommended Answers

All 3 Replies

If you have port forwarded both ports and one works fine it sounds like a problem on the webserver. Can you connect to the webserver internally using https?

Post your apache config file here as well as the output from iptables-save(run that command as root).

Please use code tags when you post the information

Hello,

thanks for your quick contact. I have read some of your posts - you are well versed in these matters I see!

The https:// was reachable on the LAN.

I have not posted the results - it works now. Why? I have no idea, but we use Magic Online here in France, and I have been battling with this problem with them for over a week. I thought the router was the problem, but that would be unusual, wouldn't it? Yesterday, they apparently brought in an expert, and the upshot is that this morning, after rebooting the router, I have the "It works!" page from apache2 from the Internet and the LAN.

I did suggest to them that it might be a setting between them and the router - we have given notice to the middle of next year, and I would not be surprised if there was a mistaken tweak somewhere...

I have sent an email to them to ask what they have done to the router settings. I don't have an an answer as yet, but I will let you know when I do.

Anyway, thank you for your time!

Kind regards

I'm glad you got it working but you're doing the right thing in finding out why it didn't work to begin with as it could likely break again in the future.

Please mark this thread as solved since you have it working now and if the connection issue comes back please update this thread and we can look it in to it. Good luck!

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.