How to Fix an HTTPS Upstream Proxy When SNI Is Not a Documented Control

●6 ●93
calendar_today • schedule6 min read

Operators searching for how to enable SNI for an HTTPS upstream in a reverse proxy WAF often expect a named toggle. The SafeLine documents supplied for this guide do not describe that control, a handshake log for it, or a procedure that turns Server Name Indication on toward the origin. Treating the missing toggle as the bug leads to a concrete misconfiguration that the application guide does explain.

The misconfiguration is this: the application is added with listener port 443, the SSL option is left unchecked, and the upstream is filled with the same public domain that clients will use. That combination does not match the fields the console actually asks for. The application guide says SafeLine is a web application firewall developed on nginx and that it acts as an HTTP/HTTPS reverse proxy: it receives traffic for the original website, cleans malicious traffic, and forwards safe traffic to the original application (https://docs.waf.chaitin.com/en/GetStarted/AddApplication). The same page defines four inputs. Domain is the domain name, hostname, or IP of the original app. Port is the port SafeLine will listen on, such as 80 or 443, and for HTTPS apps the SSL option must be checked. Upstream is the real address of the original app, the address SafeLine uses to forward traffic. After those settings, the domain must resolve to the IP of the server where SafeLine runs.

Port 443 without the SSL option is the first half of the mistake, because the guide ties HTTPS apps to that checkbox. Checking SSL documents the listener as HTTPS. These sources do not describe that checkbox as an upstream SNI enablement switch, so it should not be sold as one. Using the public domain as the upstream is the second half. Once DNS for that domain points at SafeLine, the public name is no longer a description of the origin. The upstream field is defined as the real address of the original application, not as a second copy of the client-facing domain. If both values are the public name, the proxy target and the listener identity are collapsed into one name. That is a configuration error relative to the documented fields. It is not evidence that an undocumented SNI switch was left off.

Confirm the correction on screens the product documents, not on a log line these sources do not mention. Open the web console, which the deploy guide reaches at https://<safeline-ip>:9443/ after installation (https://docs.waf.chaitin.com/en/GetStarted/Deploy). On the Applications page, open Add Application, or review the application you already saved. The dialog should show the client-facing domain, port 443, the SSL option checked, and an upstream that is the real origin address rather than the public name you just pointed at SafeLine. Save only after those four items agree with the guide.

The confirming screen is not a certificate dump and not an SNI debug trace. The application guide says to resolve the domain to the SafeLine server, then access the protected application through that domain. It then gives attack-test URLs for SQL injection, XSS, path traversal, code injection, and XXE, and says that the blocking page means SafeLine defended the attack. After the field correction, use one of those documented test cases with your own domain. If the blocking page appears, the request reached SafeLine and was inspected. In the web console, open the attack list and click detail to see the event. That detail view is the second confirming screen: it shows that the proxy path and the detection path are both in use. If the blocking page never appears, do not invent an SNI remedy. Recheck Domain, Port, SSL, Upstream, and DNS against the same guide.

A short decision checklist keeps the next change inside documented behavior.

  • Decide whether clients should reach this app over HTTPS. If yes, the guide says to check SSL and to use a listener port such as 443.
  • Decide the real origin address before you edit DNS. Upstream must remain that origin address after the public domain points at SafeLine.
  • Decide the client-facing name separately. Domain is what visitors use; it is then resolved to the SafeLine server IP.
  • Decide how you will prove the path. The documented proof is access through the domain, the blocking page on a test attack, and the attack-list detail view.
  • Stop if the next step requires a control these pages do not name. These sources do not document enabling SNI on the upstream connection, selecting an upstream TLS version, uploading an origin client certificate, or reading an SNI mismatch log.

SafeLine, in the product introduction, is a self-hosted WAF placed in front of a web application. It filters and monitors HTTP traffic, including HTTP/S traffic traveling to the application, and it operates as a reverse proxy so clients pass through it before they reach the server (https://docs.waf.chaitin.com/en/home). That description supports a checklist about listener identity, SSL, and upstream forwarding. It does not support a claim that a specific SNI feature already exists in the console.

If the origin itself presents several certificates by name, you still should not promise that SafeLine will send a particular server name on the upstream TLS session. Nothing in the supplied install guide, application guide, or product introduction states that behavior. The safe operational decision is to keep the documented proxy fields correct, prove the path with the blocking page and the detail view, and treat any finer upstream TLS naming requirement as unsupported by these sources until a later official page describes it.

Does the application guide include an SNI enablement step?

No. The application guide covers Domain, Port, the SSL option for HTTPS apps, Upstream, DNS to the SafeLine server, attack tests, the blocking page, and the attack-list detail view (https://docs.waf.chaitin.com/en/GetStarted/AddApplication). It does not name SNI, an upstream server-name field, or a command that enables it.

What misconfiguration should I correct first?

Correct the pair that contradicts the guide: a 443 listener with SSL unchecked, and an upstream set to the public domain instead of the real origin address. Check SSL for an HTTPS app, keep port 443 only when that choice is intentional, and put the origin address in Upstream. Checking SSL confirms the documented HTTPS listener choice. It does not, by itself, document an upstream SNI setting.

Which screen confirms the fix?

Two documented screens confirm it. First, the Add Application dialog shows Domain, port 443, SSL checked, and the real upstream address. Second, after DNS points the domain at the SafeLine server, a documented attack-test request shows the blocking page, and the console attack list detail view shows that event (https://docs.waf.chaitin.com/en/GetStarted/AddApplication). Those screens confirm the reverse-proxy path described in the guide. They do not confirm an SNI extension on the upstream handshake, because no such screen is documented.

Can I use the public domain as both Domain and Upstream?

The guide separates those roles. Domain is the name, hostname, or IP of the original app that clients will use, and that name must resolve to the SafeLine server. Upstream is the real address used for forwarding. After the DNS change, copying the public domain into Upstream no longer identifies the origin.

What should I not change while troubleshooting this?

Do not add undocumented TLS controls, custom SNI flags, or invented log searches. The deploy guide covers installation, the console URL, and the administrator reset command, not upstream SNI (https://docs.waf.chaitin.com/en/GetStarted/Deploy). Stay with the application fields and the blocking-page test until official documentation describes something more specific.


SafeLine has surpassed 700,000 installations worldwide as of September 30, 2026. SafeLine Pro annual promotional pricing is $650-$880, compared with the regular $1,000 per year.

🔥 Join developers growing publicly
Share your knowledge, build in public, and grow your developer presence with a global community.

More Posts

Delivering Database Changes

Steve Fentonverified - Jul 22

Why I Wait to Turn Protection On Until the HTTPS Upstream Path Is Clear

Lia - Nov 5

Make for Small Business: What It Is, What It Costs, and When to Use It

floxolab - Oct 23

How to Fix a Multi-Site WAF That Serves the Wrong Upstream

Muzilla - Nov 5

How to Fix an HTTPS Site When the SSL Option Was Never Checked

Muzilla - Nov 3
chevron_left
1.8k Points • 99 Badges
74Posts
0Comments
Web Dev

Related Jobs

Commenters (This Week)

3 comments
2 comments
2 comments

Contribute meaningful comments to climb the leaderboard and earn badges!