Custom domain stuck on "Waiting for DNS update" despite TXT/CNAME being correctly propagated externa

Question

**Project:** da9e7b1f-c9b0-46b1-bf4f-8d5677346538 (env: production)
**Service:** @animus-v2/web (c2e2cab9-9049-4450-ba88-e326bbb24fd3)
**Custom domain:** animus.eduzz.com
**DNS provider:** AWS Route 53

**Expected records (from Railway panel):**
- CNAME `animus` → `fr4br30f.up.railway.app` ✅ validated
- TXT `_railway-verify.animus` → `railway-verify=33e5abdb5028e0df6862a8649dc0015499a05d7ca1c7f1e1af7ed46171d998bc` ⚠️ stuck

**External DNS state** (Google 8.8.8.8 and Cloudflare 1.1.1.1, TTL 300):
Both CNAME and TXT resolve with the exact expected values. Propagated >24h.

**Edge:**
$ curl -sI https://animus.eduzz.com/
HTTP/1.1 404 Not Found
Server: railway-hikari
x-railway-fallback: true

TLS handshake returns wildcard `*.up.railway.app` cert (custom cert never issued).

**Things I tried (none unstuck the pipeline):**
- Redeploy + Restart of the web service
- Editing custom domain port via UI (8080 → 3000 → 8080) to force re-poll
- Confirmed no CAA record on eduzz.com blocks Let's Encrypt
- Confirmed no wildcard DNS interference (random subdomain returns NXDOMAIN)

**Likely cause:** the subdomain previously pointed to AWS CloudFront; suspect Railway's poller is stuck on a stale negative cache from before the CNAME switch.

**Ask:** could you manually re-trigger the custom-domain validation on the backend? I'd rather not remove/readd the domain since the TXT might be regenerated, which would mean another round-trip with our third-party DNS team.

Thanks!