25 days ago
Custom domains www.lcsportal.co.uk, lcsportal.co.uk and portal.lcsportal.co.uk on project “LCS Portal”, service “LCS Test Portal” (production) all show “Let’s Encrypt rate limit reached”, and retry is disabled. DNS is verified and correct. portal.lcsportal.co.uk was added at 15:11 UTC today and failed within minutes, while om-manuals.lcsportal.co.uk (another service) issued fine at 13:49. Could you clear the hold or reissue? Our live portal is down until then.
5 Replies
25 days ago
We checked the domains on that service. Ownership verification has passed on them, but their certificate requests were refused by Let's Encrypt with a rate-limit error. That limit is set by Let's Encrypt, not by us, so we can't clear it from our side. Until it resets, a reissue would be refused the same way.
Removing and re-adding the domains won't get around it either. Each new attempt counts toward the same limit, so we'd recommend holding off on that while you wait.
Separately, when we checked the most recently added subdomain, our view didn't show its CNAME record yet. It needs a CNAME pointing to the target listed for it in the service's Networking settings before a certificate can be issued once the limit resets.
Status changed to Awaiting User Response Railway • 25 days ago
Railway
We checked the domains on that service. Ownership verification has passed on them, but their certificate requests were refused by Let's Encrypt with a rate-limit error. That limit is set by Let's Encrypt, not by us, so we can't clear it from our side. Until it resets, a reissue would be refused the same way. Removing and re-adding the domains won't get around it either. Each new attempt counts toward the same limit, so we'd recommend holding off on that while you wait. Separately, when we checked the most recently added subdomain, our view didn't show its CNAME record yet. It needs a CNAME pointing to the target listed for it in the service's Networking settings before a certificate can be issued once the limit resets.
25 days ago
Thanks. The portal CNAME is in place (portal → vx56oq60.up.railway.app, and your Networking page shows it as propagated). Could a team member tell us which Let’s Encrypt limit we hit and when it resets? It’s odd that a name added at 15:11 failed at once, when other names on the same domain issued at 13:49.
Status changed to Awaiting Railway Response Railway • 25 days ago
25 days ago
The CNAME on that new subdomain now shows as propagated with the correct target on our side, so the DNS for it is in place.
All three domains are still recording the same refusal from Let's Encrypt, a rate-limit error that doesn't say which of their limits was hit or when it resets. So we can't tell you a reset time, or why a name added at 15:11 UTC was refused when others on the same domain issued at 13:49 UTC. With this error, the only fix is to wait for the limit to clear. Retrying before then, or removing and re-adding the domains, will be refused the same way.
Status changed to Awaiting User Response Railway • 25 days ago
25 days ago
We're seeing exactly the same thing this afternoon, and the timing across threads suggests
this isn't a per-hostname limit caused by individual users' attempts but something on
Railway's side (e.g. a limit on Railway's own Let's Encrypt/ACME account).
Our case:
-
Added die-pizzabaeckerei.de and www.die-pizzabaeckerei.de to a production service at
15:16 UTC today.
-
www.die-pizzabaeckerei.de reported "Let's Encrypt rate limit reached" 0.9 s after it was
created (created 15:16:25.034, failed 15:16:25.903 UTC) — before any HTTP-01 validation
could have run or failed. There were no earlier attempts for this hostname on our side.
-
The apex failed the same way at 15:21 UTC. Both now show "retryable: false".
-
Let's Encrypt has issued no certificate for this domain recently (crt.sh), so it isn't the
duplicate-certificate or certificates-per-domain limit either.
-
DNS is correct and verified: apex ALIAS → our up.railway.app target, www CNAME → its target,
no AAAA, no CAA, and /.well-known/acme-challenge/ is answered by railway-edge.
The pattern across threads today:
- 13:49 UTC — issuance still worked (other service in this thread)
- 14:32 UTC — same error reported in /questions/let-s-encrypt-tls-certificate-rate-limit-1b43df53
- 15:11 UTC — this thread's new subdomain fails within minutes
- 15:16 UTC — ours fails on the very first order
Three unrelated customers, failing on the first order within the same ~90 minutes, while Let's
Encrypt's status page shows no incident. Could someone check whether Railway's ACME account is
hitting a Let's Encrypt account-level limit (e.g. new orders per account) or whether the
certificate pipeline has an issue, and post an incident if so? "Wait for the window to pass"
doesn't help if we don't know which limit it is or whether the window applies to us at all.
Happy to share domain/project IDs privately.
Status changed to Awaiting Railway Response Railway • 25 days ago
Status changed to Awaiting User Response Railway • 25 days ago
24 days ago
Certificate issuance for lcsportal.co.uk, www.lcsportal.co.uk and portal.lcsportal.co.uk has been restarted from our side, with the domains kept as they are: the existing CNAME targets and verification records are unchanged, so there is nothing to update in your DNS. All three certificates are now valid and HTTPS is serving on each hostname. If a browser still shows the old warning, that is a cached result and a fresh connection clears it.
For die-pizzabaeckerei.de and www.die-pizzabaeckerei.de: both certificates show as valid on our side now. If either hostname still errors for you, check the DNS records listed for it in the service's Networking settings, since our last check could not confirm the current record values.
The failures came from a rate limit on our shared account with the certificate authority rather than from any of your domains, which is why no retry time was attached to the error.
Status changed to Solved dizzydes90 • 24 days ago