How-to
HTTP Methods and Required Logpush Fields
What the HTTP methods card tells you, and how to fix the "(no method)" bar by adding ClientRequestMethod to your Logpush job.
Jump to section
What the HTTP methods card shows
On every Path detail page, the HTTP methods breakdown shows how requests to that path split by method: GET, POST, PUT, etc. Colour-coded:
- Green — read-only (GET, HEAD, OPTIONS)
- Amber — mutating (POST, PUT, PATCH, DELETE)
- Red — (no method) — Logpush didn't supply a method for those requests
Why this matters
Different attacks prefer different methods. Some real-world patterns:
- Gift-card enumeration via COBilling-RedeemGiftCertificate — high POST volume from residential IPs. Legitimate gift-card redemptions are rare and evenly distributed; a burst of POSTs from 20 different ASNs in 10 minutes is almost always enumeration.
- Password-reset email spam / account enumeration — mostly POST on Account-PasswordReset from distributed residential IPs + datacentre egress. See Account-PasswordReset abuse patterns.
- Form-probing before an actual attack — scrapers load the form (GET) to extract the CSRF token template before submitting (POST). A sudden shift from mostly-GET to mostly-POST on a prescribed endpoint can precede active exploitation.
- Scanning / fuzzing — high GET volume on endpoints that shouldn't be GET-accessible. Classic directory enumeration or API probing.
Fixing "(no method)"
If the HTTP methods card on a path shows a large red (no method) bar — especially 100% — it means the Cloudflare Logpush job for your source isn't shipping the ClientRequestMethod field to Blankitt.
How to fix:
- Open your Cloudflare dashboard → select your zone → Analytics & Logs → Logpush
- Edit the Logpush job that points at Blankitt's ingest URL
- In the field selection, ensure
ClientRequestMethodis ticked - Save the job
The next Logpush batch (≤ 5 minutes) will start including the method. Historical data from before the fix will continue to show (no method) — that portion fades out naturally as it ages out of the 90-day Analytics Engine retention window.
The full required-fields list
Logpush is opt-in per field. The following fields should all be selected on your job config; missing any of them causes specific Edge features to degrade silently.
| Cloudflare field | What breaks without it |
|---|---|
EdgeStartTimestamp | Ingest fails entirely — this is required |
ClientIP | IPs page, Unique IPs column, forensic IP-discovery |
ClientASN | Offenders page, asn_spike and most other detectors |
ClientCountry | Country-scoped baselines, WorldMap |
ClientRequestPath / ClientRequestURI | Paths page, path_spike detector |
ClientRequestMethod | HTTP methods card (this article) |
ClientRequestUserAgent | UA families, ua_rotation detector |
EdgeResponseStatus | Status classes, 499_rate detector |
EdgeResponseBytes | Bytes columns everywhere |
OriginResponseDurationMs | origin_latency_spike detector, origin-latency diagnostics. SFCC renamed this from the legacy OriginResponseTime in the current catalogue; only the new name is accepted from 2026-04-22. |
CacheCacheStatus | Cache breakdown, cache_bypass detector |
EdgeColoCode | Data Centres panel |
BotScore (SFCC eCDN only) | bot_score detector |
ClientSSLProtocol | tls_weak_protocol detector |
SecurityActions (SFCC) / FirewallMatchesActions (raw Cloudflare) | challenge_solving detector |
If a detector or UI panel isn't populating as expected, the first thing to check is whether its required field is in the Logpush config.