Skip to main content

Is LaunchDarkly Down?

No — LaunchDarkly is up

Reachable from all 8 checked regions

Average response time: 107ms

Last checked · checks run every 6 hours

Official status page: https://status.launchdarkly.com

Flag Updates via Polling API delayed

minor investigating

We're experiencing a service degradation which can result in Flag updates via the Polling API to be delayed.

Components the vendor reports as affected: Polling API Degraded Performance

A component listed here does not necessarily mean a full outage — large providers routinely report maintenance on individual locations. It only affects the verdict above when the vendor marks it major or critical and our own checks also saw LaunchDarkly fail from enough regions to rule out probe noise.

LaunchDarkly uptime

100%
Last 7 days
100%
Last 30 days
100%
Last 90 days
157ms
Avg response, 30 days

Measured from multiple regions every 6 hours. Percentages count only checks that returned an availability answer — 19 days measured so far. A dash means that window does not yet hold enough measured days to publish a figure.

30-day history

19-day clean streak
Jul 19: no data
Jul 20: no data
Jul 21: no data
Jul 22: no data
Jul 23: no data
Jul 24: no data
Jul 25: no data
Jul 26: no data
Jul 27: no data
Jul 28: no data
Jul 29: no data
Jul 30: 100.00% uptime, 40 checks
Jul 31: 100.00% uptime, 32 checks
Aug 1: 100.00% uptime, 32 checks
Aug 2: 100.00% uptime, 32 checks
Aug 3: 100.00% uptime, 32 checks
Aug 4: 100.00% uptime, 32 checks
Aug 5: 100.00% uptime, 40 checks
Aug 6: 100.00% uptime, 48 checks
Aug 7: 100.00% uptime, 32 checks
Aug 8: 100.00% uptime, 32 checks
Aug 9: 100.00% uptime, 32 checks
Aug 10: 100.00% uptime, 32 checks
Aug 11: 100.00% uptime, 32 checks
Aug 12: 100.00% uptime, 40 checks
Aug 13: 100.00% uptime, 32 checks
Aug 14: 100.00% uptime, 32 checks
Aug 15: 100.00% uptime, 32 checks
Aug 16: 100.00% uptime, 31 checks
Aug 17: 100.00% uptime, 24 checks
Jul 19 Today
No downtime Partial Downtime Not measurable No data

Reachability by region

Each region runs its own request from a different part of the world. A service can be up for one continent and down for another, which is usually the first sign of a routing or CDN problem.

ams
104ms
DNS 0ms TCP 1ms TLS 24ms TTFB 62ms
arn
217ms
DNS 1ms TCP 0ms TLS 56ms TTFB 110ms
bom
107ms
DNS 1ms TCP 2ms TLS 23ms TTFB 71ms
nrt
52ms
DNS 0ms TCP 2ms TLS 10ms TTFB 31ms
ord
151ms
DNS 0ms TCP 11ms TLS 16ms TTFB 77ms
sin
76ms
DNS 1ms TCP 1ms TLS 13ms TTFB 54ms
sjc
52ms
DNS 0ms TCP 1ms TLS 9ms TTFB 38ms
syd
45ms
DNS 0ms TCP 0ms TLS 10ms TTFB 26ms
yyz
107ms
DNS 0ms TCP 0ms TLS 23ms TTFB 65ms

What LaunchDarkly does

LaunchDarkly is a feature management platform that decides, at runtime, which users see which version of a feature. Engineering teams use it to release code behind flags, roll changes out gradually and switch a broken feature off without deploying. That makes it a live dependency of the application rather than a build-time tool.

What an outage looks like

Flag changes made in the LaunchDarkly dashboard stop reaching applications, so a rollout appears to save but never takes effect for users. Applications starting up fresh may fall back to the default values compiled into their code, which can turn features off for everyone at once or expose ones meant to stay hidden. The web app itself may fail to load or save targeting rules.

What to do about it

Check status.launchdarkly.com, which separates the Flag Delivery Network from the web app, the streaming and polling APIs, and experimentation, because flags already delivered can keep working while the dashboard is unavailable. SDKs cache their last known flag values, so running services usually continue on the previous configuration. Delay restarts and deploys until delivery recovers, since a restart is what forces a service onto defaults.

Is it down for everyone, or just you?

If this page says LaunchDarkly is up but it is not loading for you, the problem is between you and them. Run a check against any URL from all 18 regions to find out where it breaks.

Test it yourself

Related services

Datadog
Datadog
Sentry
Sentry
GitHub
GitHub
PagerDuty
PagerDuty

LaunchDarkly outage FAQ

Do my feature flags still work if LaunchDarkly is down?
Generally yes for services that are already running. LaunchDarkly SDKs hold the last flag values they received in memory, so an application that was connected before the incident keeps evaluating flags normally even when it cannot reach the service. What stops is change: new flag updates do not arrive, so anything you toggle during the outage will not take effect until delivery recovers.
What happens when a service restarts during a LaunchDarkly outage?
It loses the cached values and falls back to the defaults written into your code. This is the failure mode that turns a LaunchDarkly incident into a user-visible one, because a routine deploy or autoscaling event can flip a whole fleet onto default behaviour mid-incident. Holding deploys and restarts until the Flag Delivery Network is healthy is the single most useful thing to do.
Is the LaunchDarkly dashboard down or is flag delivery affected?
They are separate components and fail independently. The status page lists the web app apart from the Flag Delivery Network and the streaming, polling and edge APIs, so a dashboard you cannot log into does not necessarily mean your applications are affected. Check which component is degraded before assuming production is at risk, because the two incidents call for very different responses.
Can I turn off a broken feature while LaunchDarkly is unavailable?
Not through LaunchDarkly, which is why this timing hurts. If flag delivery is degraded, a kill switch you flip in the dashboard will not reach running applications. Fall back to whatever controls you own directly, such as a configuration change, a load balancer rule or a rollback of the deploy that introduced the feature, and re-apply the flag change once delivery returns.
Does a LaunchDarkly outage affect experiments and metrics?
Yes, and it can distort results beyond the incident window. Experimentation is tracked as its own component, and when event delivery is degraded the exposures and conversions recorded during that period may be incomplete. Treat any experiment running across an outage with suspicion, and check whether the affected days should be excluded before acting on the numbers.

How we measure this

  • We request LaunchDarkly's public endpoint every 6 hours from Fly.io regions across six continents — 8 of them answered the most recent check.
  • A region counts as down only when it gets no usable HTTP response. A 403 or 429 means the origin answered and refused us, which we report as blocked, never as an outage.
  • A single failing region is treated as probe noise. We only change the verdict when two consecutive cycles agree.
  • Response times average only the regions that actually served the page, so a timeout never inflates the number.
  • Where LaunchDarkly publishes an official status feed we read it too. An all-clear from the vendor can soften an unconfirmed degradation; a vendor-declared outage only worsens our verdict when our own checks corroborate it.

Get alerted when LaunchDarkly goes down

This page refreshes every 6 hours. Your own monitors run as often as every 30 seconds, from the same 18 regions, and tell you the moment something breaks.

Start Free Monitoring
Free plan available No credit card required