Stale DNS cache prevents custom domain validation after Cloudflare migration
leobrant23
HOBBYOP

11 days ago

Hi! I added a custom domain www.logistical-solution.com on service logic-tms-frontend. The DNS public resolution is correct (CNAME → 2g08wmeq.up.railway.app), confirmed via Cloudflare/Google/Quad9 resolvers. But your "Configure DNS Records" panel keeps showing Current: "zhs.zohosites.com", which is the OLD value from before I migrated DNS to Cloudflare. I've already removed and re-added the domain twice. This looks like a stale internal resolver cache. Could you force a DNS refresh on your end? Thanks.

UPDATE: I also tried redeploying the frontend service — no change. The Networking panel still shows Current: "zhs.zohosites.com" for the www CNAME despite the public DNS being correct (verified via dig @1.1.1.1, @8.8.8.8, @9.9.9.9 — all return CNAME → 2g08wmeq.up.railway.app → 66.33.22.190). Same problem for the apex domain logistical-solution.com which shows "Waiting for DNS update" with no setup-issues link visible.

Tried so far:

  • Removed and re-added both custom domains (twice)
  • Migrated DNS authority from OpenSRS to Cloudflare (CNAME flattening for apex)
  • Triggered manual redeploy
  • Confirmed TXT verify records are green ✓ on both

The api.logistical-solution.com domain on a DIFFERENT service (logic-tms-backend) validated correctly using the same Cloudflare setup. Only this service's domains are stuck.

This appears identical to a previously reported issue: thread "Custom Domain Edge Routing Failure" (1 month ago, PINNED SOLUTION). Same symptoms — custom domain stuck in VALIDATING_OWNERSHIP, edge returns 404, recreate-and-re-add has no effect. The moderator's reply there said the DNS records were missing, but I've triple-confirmed mine are correct (public DNS resolves via 4 different resolvers; TXT verify is green ✓ on both domains; CNAMEs unproxied on Cloudflare). Could you check the edge ingress config sync for service logic-tms-frontend specifically? My request IDs are visible in the dashboard.

$10 Bounty

2 Replies

Status changed to Open Railway 11 days ago


zx1234r
PRO

11 days ago

Hey leobrant23,

This is clearly a stale internal DNS cache on Railway's edge/ingress layer - your public DNS is correct and verified across multiple resolvers, so the issue is on Railway's side, not yours.

Key signal: api.logistical-solution.com on a different service validated fine with the same Cloudflare setup, which confirms your DNS config is correct and isolates the problem to the logic-tms-frontend service specifically.

Since you've already exhausted all user-side options (remove/re-add twice, redeploy, TXT records green on both domains), this needs Railway staff to:

Force a DNS cache flush for www.logistical-solution.com and logistical-solution.com on their internal resolver/ingress layer

Re-sync the edge ingress config for service logic-tms-frontend specifically

Check why the apex domain shows "Waiting for DNS update" with no setup-issues link - that's an abnormal UI state


It may take up to a few hours (maybe even a day) for Railway to validate the DNS and issue certificates. If one isn't issued by then, try removing the domain from Railway and add it back after ~10-15 mins. Update DNS records as necessary.


Welcome!

Sign in to your Railway account to join the conversation.

Loading...