Automatic build runs into error

[brankovics (Balázs Brankovics)]

I have set up docker repo (brankovics/grabb) that should automatically build from my GitHub repo, both repositories are public. The automatic build appears to crash when the fully built image is being pushed:

{"status":"Pushing","progressDetail":{"current":61105,"total":61105},"progress":"[==================================================\u003e]  61.1 kB/61.1 kB","id":"d532e9146af0"}
{"errorDetail":{"message":"Head https://dseasb33srnrn.cloudfront.net/registry-v2/docker/registry/v2/blobs/sha256/03/032941ff698ab28442116cf383d70cb209dfc10e00b5b43f9cfb06ac220707f4/data?Expires=1447764061\u0026Signature=jHf4tNbdhQgTtnskAz3dY869MuWissqE8FKm10ndYhIbNlv~O-DHzfOy0QtLEoRSGv1diuWynBKAcQdNccxu5XP3AF2bLWy2ucWqDf7AylJsx8B239fu98zOObuEHFSH1mmG85N4L2GWjZM3MfFaEH-jmEzorLn~Tc25bZ1TvJ8_\u0026Key-Pair-Id=APKAJECH5M7VWIS5YZ6Q: dial tcp 54.239.152.44:443: i/o timeout"},"error":"Head https://dseasb33srnrn.cloudfront.net/registry-v2/docker/registry/v2/blobs/sha256/03/032941ff698ab28442116cf383d70cb209dfc10e00b5b43f9cfb06ac220707f4/data?Expires=1447764061\u0026Signature=jHf4tNbdhQgTtnskAz3dY869MuWissqE8FKm10ndYhIbNlv~O-DHzfOy0QtLEoRSGv1diuWynBKAcQdNccxu5XP3AF2bLWy2ucWqDf7AylJsx8B239fu98zOObuEHFSH1mmG85N4L2GWjZM3MfFaEH-jmEzorLn~Tc25bZ1TvJ8_\u0026Key-Pair-Id=APKAJECH5M7VWIS5YZ6Q: dial tcp 54.239.152.44:443: i/o timeout"}

I was guessing that something is wrong with the deploy keys, so I did the following:

1. I deleted all the deploy keys of my GitHub repo, since Docker should automatically create new ones when using public repos. (At least that is what I understood.) Triggered a build, which failed.
2. Added a new deploy key by copying it from the docker repo build settings. Triggered a build, which failed.
3. Deleted the previous key and created one with write access. Triggered a build, which failed.

I have tested locally whether the github repo url can be used for a successful build and it worked.
Could someone help me solve this problem?

[thefynx (Levi Smith)]

Lots of others having this error, myself included.

Push failed on automated build

Hello, I created an automated repo recently with a Dockerfile and README.md. After several tries to build the images, I’ve got alternativly these errors: Push failed: {"errorDetail":{"message":"Head https://dseasb33srnrn.cloudfront.net/registry-v2/***Key-Pair-Id=****: dial tcp 54.230.195.162:443: i/o timeout"}} Or: Build process failed: Authentication is required. I manage to build the image flawlessly in several local environment. Is there something I can do?

[brankovics (Balázs Brankovics)]

I have managed to solve the problem. I will list what I have done in order to make it truly reproducible:

1. Deleted Docker form the services listed from the GitHub repo.
2. Deleted the linking of GitHub and Docker accounts on the Docker Hub end.
3. Re-established the link. Chose for the recommended one.
4. Deleted the deploy keys from the Github repo.
5. Deleted Docker repo.
6. Re-created Docker repo as an automatic build and public.
7. (At the Docker build settings the deploy key disappeared.) At the build settings I linked my repo with the docker repo that I used with FROM in the docker file.
8. Also added the built setting that from the master branch it should create an image with latest tag. Then triggered a build.

I believe that the last two steps could have been the crucial ones. (Although the fact that the deploy key disappeared could also be important.)

[allenlai]

We are aware of the issue and our Engineers are working feverishly on a fix. We don’t have a timeline to share for the fix at this time. However, we recommend that you go to the Docker Status Page to get the latest update. You can subscribe to the incident and be automatically notified when an update is available.

We apologize for the inconvenience and thank you for your patience during this matter.