Dev Tools Status
Everything is up.
Developer platforms are made of parts that fail independently, and knowing which part is down tells you whether to wait or work around it. Git hosting, the web interface, the API, package registries, and CI runners each have their own failure modes. A push can succeed while pipelines sit queued for an hour, because the thing that broke is runner capacity, not the repository.
The pattern worth knowing is that build and deploy systems are far more fragile than the edge that serves already-published output. On hosting platforms this means an incident usually blocks new deployments while every live site keeps serving traffic normally. Registries invert that: when a package registry degrades, local work with a warm cache continues while every CI pipeline in the world starts failing within minutes.
We probe the API or an application path rather than a marketing page wherever one exists, because a company's homepage staying up says nothing about whether the thing you use works.
- up
- slow
- down
- no reading
CircleCIUp
Docker HubUp
ExpoUp
Fly.ioUp
GitHubUp
GitLabUp
Grafana CloudUp
HerokuUp
LinearUp
NetlifyUp
npmUp
PostmanUp
RailwayUp
RenderUp
SentryUp
VercelUp
Common questions about developer tools
Queued jobs are usually capacity rather than an outage, and most providers do not open an incident for queue depth. Check whether jobs are failing or merely waiting. Waiting with no error points at shared runner contention, which typically clears on its own and is the most common reason CI feels broken while everything reports healthy.
They are separate systems. Git over SSH and HTTPS, the web interface, the API, and the package registry are all reported separately by most providers precisely because any one of them can fail alone. A verdict here covers the endpoint we check, so the linked official status page is worth reading for a per-component breakdown.
Retry once, since build steps fail on transient network errors more often than people expect. If it fails identically twice with the same error at the same step, that points at your build rather than the platform. Platform incidents usually fail at a different stage each time, or hang before your build starts at all.