LaunchDarkly Login
The LaunchDarkly dashboard is at app.launchdarkly.com. If you are on their US federal instance, it is app.launchdarkly.us instead, and credentials do not carry across. That is the whole answer; the rest of this page is for when signing in does not work.
Worth knowing who wrote this: we're ShipSilently, a feature flag service that competes with LaunchDarkly. We've kept this page to things you can verify, with sources at the bottom, because a page that shades the facts is worth less to us than one you'd send to a colleague. Found something wrong or out of date? Tell us and we'll correct it.
Not affiliated with or endorsed by LaunchDarkly. LaunchDarkly is a trademark of its owner, used here to refer to their product. For official documentation see launchdarkly.com/docs.
Which URL you need
| Instance | Sign in at | Status page |
|---|---|---|
| Commercial The default for almost everyone. | app.launchdarkly.com | status.launchdarkly.com |
| US federal Separate deployment, separate accounts. | app.launchdarkly.us | status.launchdarkly.us |
The two are genuinely separate systems. An account on one does not exist on the other, and neither do its SDK keys or API tokens.
When SSO is in the way
LaunchDarkly implements single sign-on over SAML 2.0, and there are two ways a session can start. Service-provider-initiated means you go to the LaunchDarkly login page and it hands you off to your identity provider. Identity-provider-initiated means you start from your company's app portal (Okta, Entra, Google Workspace) and click through from there. Which one your organisation uses is an administrator's choice, and it changes what a working login looks like.
Three failure modes account for most SSO login tickets:
- Test-drive mode. After a valid SAML configuration is accepted, SSO sits in a verification state where the login screen still asks for a password. Users read that as "SSO is broken". It isn't, it just hasn't been switched on for everyone yet.
- SSO-provisioned users needing a password reset. LaunchDarkly's own documentation calls this out: accounts created through SSO provisioning have to go through a password reset before they can sign in. Run the reset from the login page before escalating.
- SAML assertion errors. These surface as an error string on the callback rather than a generic failure, which is helpful, because LaunchDarkly documents what each one means. Usual suspects are a mismatched ACS URL or entity ID, a missing attribute in the assertion, an expired IdP signing certificate, or clock drift.
Only the first is something you can fix yourself. The other two need whoever owns the IdP configuration, so the fastest route is usually to screenshot the exact error and send it to them rather than retrying.
When the dashboard itself is down
Worth separating two situations that feel identical at the login screen. Sometimes you can't sign in. Sometimes nobody can, because the web application is unavailable, which has happened, including a window in July 2026 when the dashboard was down at the same time as flag evaluations were erroring. We transcribed that sequence from their status page on our LaunchDarkly status page.
The useful thing to know is that the REST API is a different surface from the dashboard, and it has stayed up through dashboard incidents. If you hold an access token, you can turn a flag off without the UI:
curl -X PATCH \
https://app.launchdarkly.com/api/v2/flags/default/my-flag-key \
-H "Authorization: $LD_ACCESS_TOKEN" \
-H "Content-Type: application/json; domain-model=launchdarkly.semanticpatch" \
-d '{
"environmentKey": "production",
"instructions": [{ "kind": "turnFlagOff" }]
}'Store a token with write access somewhere your on-call can reach it, and test the call once while everything is healthy. A break-glass path you have never run is not a break-glass path. The API page covers token types and the semantic patch format this uses.
One thing to check afterwards. Following a July 2026 incident, LaunchDarkly
told customers that server-side SDKs and Relay Proxies which had logged auth errors would
not reconnect on their own and needed restarting. So a green status page doesn't
automatically mean your services are receiving flag updates again. Searching your logs for
giving up permanently is the quick check.
FAQ
What is the LaunchDarkly login URL?
app.launchdarkly.com for the commercial service. Customers on the US federal instance sign in at app.launchdarkly.us, which is a separate deployment with separate accounts, separate SDK endpoints and its own status page at status.launchdarkly.us.
Why does LaunchDarkly ask for my password when my company uses SSO?
Most often because SSO is still in test-drive mode. When LaunchDarkly accepts a SAML configuration it does not switch everyone over immediately; the login screen keeps using password authentication while administrators verify the integration through their identity provider. Until an admin enables it fully, password login is expected behaviour.
I was provisioned through SSO and cannot sign in. What now?
LaunchDarkly's documentation notes that users provisioned through SSO are required to reset their password before they can sign in. Run the password reset flow from the login screen first; if that does not clear it, the issue is usually on the identity provider side and needs an administrator.
What do SAML assertion errors on login mean?
Almost always a mismatch between what your identity provider is sending and what LaunchDarkly expects: a wrong ACS URL or entity ID, a missing required attribute, a clock skew between systems, or an expired signing certificate. LaunchDarkly publishes a support article listing the specific SAML error strings and their causes, linked in the sources below.
Can I still turn off a feature flag if the LaunchDarkly dashboard is down?
Not through the UI, but the REST API is a separate surface and has stayed reachable through past dashboard incidents. If you have an access token, you can patch a flag with a single curl call. Setting that up before you need it is the point, during an incident is a bad time to discover you have no token.
Is LaunchDarkly down or is it just my login?
Check status.launchdarkly.com (or status.launchdarkly.us for federal). Their dashboard has had unavailability incidents that were unrelated to flag delivery, and, more awkwardly, at least one window where both were affected at once. We keep an independent incident summary on our LaunchDarkly status page.
More on LaunchDarkly
Not a customer yet, and comparing?
If you landed here while evaluating rather than signing in: ShipSilently is a flat $49/month feature flag service that runs on any runtime. We are missing things LaunchDarkly has, experimentation, SCIM and custom RBAC among them, and our comparison page lists them plainly next to where we win.
$49/month is the 2026 price. List goes to $89/month on January 1, 2027, and subscriptions started before December 31, 2026 stay at $49.
Sources
- Logging in with SSO: launchdarkly.com/docs/home/getting-started/login-sso
- Single sign-on overview and SAML configuration: launchdarkly.com/docs/home/account/sso and /docs/home/account/saml (test-drive mode, enabling SSO, password reset for SSO-provisioned users)
- SSO SAML error messages: support.launchdarkly.com
- Service status: status.launchdarkly.com, federal instance status.launchdarkly.us
Checked against the linked sources on 2026-07-25. LaunchDarkly changes its docs and pricing without notice, so treat their pages as authoritative over ours and email us when we drift.