How to Diagnose a Traffic Surge When Your WAF Sees Almost Nothing

●8 ●102
calendar_today ago • schedule4 min read

A sudden rise in login requests can signal credential stuffing: attackers automate attempts using previously stolen username and password pairs, hoping that users reused them. The immediate question is how to tell an attack surge apart from a legitimate traffic spike. But before interpreting the traffic, verify that your security controls can actually see it.

One concrete misconfiguration can undermine the entire investigation: the public application hostname still resolves to the origin server instead of the SafeLine server. The application remains reachable, its access logs grow, and the WAF's attack list provides little evidence because requests are bypassing it.

This walkthrough describes a diagnostic scenario, not a reported incident.

Start with the visibility mismatch

SafeLine operates as an HTTP/HTTPS reverse proxy: it receives application traffic and forwards permitted traffic to the upstream application. Its documented application setup requires configuring Domain, Port, and Upstream, then resolving the application domain to the SafeLine server's IP address. Official application configuration guide

Suppose your application logs show a sharp increase in login submissions, but you cannot find corresponding evidence in your WAF investigation. An empty attack list does not establish that the requests are legitimate. Requests might bypass the WAF, or they might reach it without triggering an attack detection.

First establish the request path. Check public DNS answers for the affected hostname, including both IPv4 and IPv6 records where present. Compare those answers with the intended SafeLine listener address. Then review the application's entry in the SafeLine Applications page: the domain must match the hostname being investigated, the listening port must match the service, and Upstream must identify the actual application server. These fields and the Applications page are documented in the configuration guide.

Fix the routing error and collect proof

For this specific misconfiguration, correct the public DNS record that still points directly to the origin so it points to the SafeLine server, following the documented deployment model. Keep the upstream configured as the application's real address. Account for DNS caching before assuming every client has switched paths.

Confirm the correction with two different observations:

  1. Send a normal request through the public hostname and confirm that the application still responds correctly. Correlate its timestamp and a harmless unique query value with your application access log.
  2. On a staging application or an explicitly authorized test target, use a suitable test request from the official guide. Confirm that it produces the documented blocking page, then open the SafeLine attack list and click "detail" to inspect the matching event. The guide documents both the blocking page and this console workflow. Official test and verification instructions

The matching attack detail is the screen that confirms the tested request reached SafeLine and triggered protection after the routing correction. It does not prove that every hostname, address family, or alternate entry point follows the same route. Check those separately, and review whether public clients can still reach the origin directly.

Now classify the login surge

Once visibility is established, compare a representative surge window with an ordinary window using application access and authentication logs. These are investigation inputs from your application, not promised SafeLine log fields.

Credential stuffing becomes more plausible when repeated login attempts span many accounts, failures dominate, and requests repeatedly target authentication endpoints with little surrounding browsing. Successful authentication followed by unfamiliar account changes deserves immediate account takeover investigation.

A legitimate spike may include more varied navigation, expected campaign timing, and successful customer activity. None of those signals proves legitimacy by itself. Attackers can distribute requests across addresses and imitate browser behavior; legitimate users can share network addresses.

Make decisions from combined evidence. Investigate suspicious successful logins, preserve relevant records, and assess authentication controls. SafeLine documents IP-based rate limiting, but shared networks and distributed attackers limit what an IP-based control alone can establish. Official capability overview

Does an empty SafeLine attack list mean the surge is safe?

No. Establish whether requests traverse the WAF, then examine application behavior. A credential-stuffing attempt can use ordinary-looking login requests without an injection payload.

Which screen confirms this routing fix?

The SafeLine attack detail for your authorized test, matched to its timing and target, confirms that the tested path reaches protection. Pair that evidence with a successful normal request to verify application availability.

Should I block the busiest IP address immediately?

Check its request behavior and authentication outcomes first. A shared network can concentrate legitimate users, while attackers can spread attempts across many addresses. Request volume alone is insufficient.

What if suspicious logins continue after the fix?

Continue the account-level investigation. A confirmed WAF path establishes visibility for the tested route; it does not establish that credentials are secure. Review suspicious sessions, affected accounts, and authentication protections using your application's evidence.


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.

Want more #security & #waf deep dives?

Join 5,973+ developers sharing real benchmarks, code breakdowns, and projects. Free forever.

More Posts

How to investigate LLM-driven login bots when traffic bypasses your WAF

Muzilla - Nov 14

Stop Implementing Authentication Inside Containers on Kubernetes

Alexandre Vazquez - Jul 25

Why I Check Login Behavior Before Enabling WAF Protection During a Surge

Lia - Nov 20

How to Block Credential Stuffing from Residential Proxy Networks: Fix the Traffic Path First

Muzilla - Nov 14

How to Stop Password Spraying Against an Admin Panel When Traffic Bypasses Your WAF

Muzilla - Nov 18
chevron_left
1.9k Points • 110 Badges
82Posts
0Comments
Web Dev

Related Jobs

View all jobs →

Commenters (This Week)

5 comments
1 comment
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!