hosting

PUT /v1/sites/{siteId}/cdn/security

Change a site's CDN security and bot settings.

Semua titik akhir hosting

Pengesahan

Hantar kunci API sebagai token pembawa. Kunci tersebut mesti membawa kebenaran hosting.cdn.manage; kunci yang tidak mempunyainya akan ditolak dengan 403, bukan 404.

Titik akhir ini tidak memerlukan id organisasi. Kunci anda telah mengenal pasti organisasi kepunyaannya, dan respons dis skopkan kepadanya.

Cuba

Gantikan apa sahaja di dalam kurungan sudut dengan nilai anda sendiri, dan pemegang tempat kunci dengan kunci dari papan pemuka anda.

curl -X PUT https://api.zinndigital.com/v1/sites/{siteId}/cdn/security \
  -H "Authorization: Bearer zdk_live_…" \
  -H "Content-Type: application/json" \
  -d '{  }'

Sudah log masuk? Konsol API dalam papan pemuka anda mengisi id organisasi sebenar dan kunci anda sendiri, serta menjalankan permintaan terhadap API langsung supaya anda boleh melihat respons sebenar. Buka penamat ini dalam konsol API

Butiran

A **merge**: only the fields present are changed. A full-object write would turn a customer's AI-crawler policy off every time they changed their security level. The body carries settings and nothing else — `additionalProperties: false`, for the reason `PUT /v1/sites/{siteId}/cdn/settings` gives: these are pooled CDN accounts and a Cloudflare zone resolves by name alone, so a body key named `domain` or `zone_id` must be refused rather than discarded. The result is read back from the provider rather than echoed, because the provider silently ignores a bot control a zone's plan does not include — "we asked" and "it changed" are different facts. Requires `hosting.cdn.manage`. **Bot management is a sold entitlement (`protection_tier`).** Turning `bot_fight_mode` on, or setting `ai_bots_protection`/`crawler_protection` to `block` or `managed_challenge`, needs a plan that includes it and otherwise answers a plan refusal naming the tier held and the tier needed. Three things are deliberately NOT gated. Turning any control **off** is free on every plan — a customer whose plan changed must always be able to reach the floor rather than be stranded with a setting they cannot undo. `allow` is an explicit relaxation, not protection, so it is free too. And `security_level: under_attack` is free on every plan, for ever: it is the emergency DDoS response, and refusing it to a customer under active attack because of their plan is not a product decision this platform will make.

Parameter

NamaJenisDiperlukanApakah ia
siteId (path)UuidYaSite ID (UUIDv7).

Badan permintaan

NamaJenisDiperlukanApakah ia
security_levelCdnSecurityLevelTidakThe provider's own security-level vocabulary, passed through. **Not ordered the way it reads** — `essentially_off` sits between `off` and `low` — so nothing may treat it as a sc…
bot_fight_modebooleanTidak
ai_bots_protectionCdnAiBotModeTidakHow the edge treats AI crawlers. `disabled` leaves them alone. This is the control on the screen most able to cost a site traffic, in either direction: blocking AI answer engine…
crawler_protectionCdnAiBotModeTidakHow the edge treats AI crawlers. `disabled` leaves them alone. This is the control on the screen most able to cost a site traffic, in either direction: blocking AI answer engine…
always_use_httpsbooleanTidak
automatic_https_rewritesbooleanTidak
opportunistic_encryptionbooleanTidak
min_tls_versionCdnMinTlsVersionTidakThe oldest TLS version the edge will negotiate. A string — `1.10` is not a version.

Respons

NamaJenisDiperlukanApakah ia
site_idUuidYaUUIDv7 identifier — sortable by creation time (docs/02 §8).
supportedbooleanYaWhether the CDN serving this site can express these settings at all. `false` for a provider with no equivalent vocabulary; the other fields then carry defaults and a client shou…
protection_tierstringYaThe protection level this organisation's plan sells (`protection_tier`). Absent from a plan means the floor — every site with a CDN already has the provider's baseline protectio…
bot_management_includedbooleanYaWhether that tier reaches bot management — enabling `bot_fight_mode`, or setting `ai_bots_protection`/`crawler_protection` to `block` or `managed_challenge`. Derived server-side…
security_levelCdnSecurityLevelYaThe provider's own security-level vocabulary, passed through. **Not ordered the way it reads** — `essentially_off` sits between `off` and `low` — so nothing may treat it as a sc…
bot_fight_modebooleanYaChallenge traffic the provider classifies as automated. Read from the provider's bot-management resource, not from a zone setting — the zone setting of that name does not exist.
ai_bots_protectionCdnAiBotModeYaHow the edge treats AI crawlers. `disabled` leaves them alone. This is the control on the screen most able to cost a site traffic, in either direction: blocking AI answer engine…
crawler_protectionCdnAiBotModeYaHow the edge treats AI crawlers. `disabled` leaves them alone. This is the control on the screen most able to cost a site traffic, in either direction: blocking AI answer engine…
always_use_httpsbooleanYaRedirect every plain-HTTP request to HTTPS at the edge.
automatic_https_rewritesbooleanYaRewrite insecure sub-resource URLs in HTML to HTTPS where possible.
opportunistic_encryptionbooleanYaAdvertise HTTP/2 over TLS to clients that arrive over plain HTTP.
min_tls_versionCdnMinTlsVersionYaThe oldest TLS version the edge will negotiate. A string — `1.10` is not a version.

Ralat yang boleh dikembalikan oleh titik akhir ini

401 · 403 · 404 · 422 · 429 · 503