My workflow says that github actions is down but the status is green

When running github actions i get:

error: failed to solve: GitHub Actions is temporarily unavailable. Please visit for the status of our services.

Error: buildx failed with: error: failed to solve: GitHub Actions is temporarily unavailable. Please visit for the status of our services.

But when looking on github status i see that everything is green. How can this be ?


I am having the same issue…

I am not sure if it’s related, but we’re getting intermittent getCacheEntry failed: Cache service responded with 503 errors when running setup-* with a cache parameter. And it seems that we’re not the only ones. GitHub Status is all green, though.


yes I think it is related because I am getting the error during the cache steps too

Also facing these issues and have been since this morning. Workflows have worked once or twice since this morning (~7AM BST). Does appear to be related to the cache.

status pages are lie. I have never see one it’s working.


Be nice to at least have them acknowledge the issue.

Hi @david-gang ,

Thanks for reporting this. We are checking it internally and will update you shortly about the current status.

Hi there,

Quick update! There is a minor interruption in the service and our team is working on it. We are sorry for the inconvenience.

You may check the current status or subscribe for the incident on

1 Like

Hi @Akash1134 ,

This minor interruption as you call it may prevent people from deploying docker images to production. So this should be dealt with all seriousness because for us developers it is a major and not minor interruption.

Besides this the fact that it took you 4 hours to identify that there is an issue is also not a merit and this should be fixed from your side.



it is super cute that you think activity in this thread indicates the timeline of literally anything that isn’t activity in this thread.

status pages are edited manually so that false positives don’t show up.

confirmation of outage and the exact nature of the outage must be known before the page can be updated with meaningful information.

also, not just anyone has access to edit those static pages. there are likely some approvals, as stupid as that sounds.

also, do you really need a status page to tell you something isn’t working?

On the contrary.
I hope they have some monitoring infrastructure which tells them if their services are functional and don’t wait for the customers to tell them that something is broken.
On the other hand from an organizational standpoint it makes sense that the status dashboard is updated by just a small number of people and they are most probably located in the United states, so they updated the dashboard when they started to work. So this is an interesting point.