Buildx: unexpected status: 401 Unauthorized
Hi All,
So I saw this was posted before but without a proper solution or breakdown why this happens. https://github.com/docker/buildx/issues/139#issuecomment-660718429
I'm also having this issue from time to time and its really unclear what it triggers.
First I do the docker login and that goes well, after that I start the buildx for the 2 images but I still get the 401 error in the end as you can see. I'm using it on Mac Docker Desktop 2.3.2.0 Edge. if you need additional information please let me know.
docker login
Authenticating with existing credentials...
Login Succeeded
docker buildx build -f Docker/alpine-cloud/api/Dockerfile --platform linux/amd64,linux/arm/v7 -t blabla1337/skf-api:dev --push .
[+] Building 1483.8s (38/38) FINISHED
=> [internal] load build definition from Dockerfile 0.1s
=> => transferring dockerfile: 32B 0.0s
=> [internal] load .dockerignore 0.1s
=> => transferring context: 2B 0.0s
=> [linux/amd64 internal] load metadata for docker.io/library/python:3.7 9.2s
=> [linux/arm/v7 internal] load metadata for docker.io/library/python:3.7 11.5s
=> [internal] load build context 89.0s
=> => transferring context: 967.06MB 81.4s
=> [linux/arm/v7 stage-1 1/16] FROM docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> => resolve docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> [linux/amd64 stage-1 1/16] FROM docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> CACHED [linux/amd64 stage-1 2/16] RUN apt update && apt install -y vim libblas-dev libblas-dev liblapack-dev libatlas-base-dev gfortran cython 0.0s
=> CACHED [linux/amd64 stage-1 3/16] RUN groupadd --gid 1000 user_api && useradd --uid 1000 --gid user_api -m user_api && mkdir -p /home/user_api 0.0s
=> [linux/amd64 stage-1 4/16] ADD ./ /home/user_api/ 27.8s
=> CACHED [linux/arm/v7 stage-1 2/16] RUN apt update && apt install -y vim libblas-dev libblas-dev liblapack-dev libatlas-base-dev gfortran cython 0.0s
=> CACHED [linux/arm/v7 stage-1 3/16] RUN groupadd --gid 1000 user_api && useradd --uid 1000 --gid user_api -m user_api && mkdir -p /home/user_api 0.0s
=> CACHED [linux/arm/v7 stage-1 4/16] ADD ./ /home/user_api/ 0.0s
=> CACHED [linux/arm/v7 stage-1 5/16] COPY ./Docker/alpine-cloud/api/entrypoint.sh /home/user_api/entrypoint.sh 0.0s
=> CACHED [linux/arm/v7 stage-1 6/16] RUN touch /home/user_api/skf/db/db.sqlite 0.0s
=> CACHED [linux/arm/v7 stage-1 7/16] RUN chmod a+rw /home/user_api/skf/db /home/user_api/skf/db/* 0.0s
=> CACHED [linux/arm/v7 stage-1 8/16] WORKDIR /home/user_api/ 0.0s
=> CACHED [linux/arm/v7 stage-1 9/16] RUN mkdir .kube 0.0s
=> CACHED [linux/arm/v7 stage-1 10/16] RUN mkdir .config 0.0s
=> CACHED [linux/arm/v7 stage-1 11/16] RUN mkdir .config/pip 0.0s
=> CACHED [linux/arm/v7 stage-1 12/16] RUN echo "[global]" >> .config/pip/pip.conf 0.0s
=> CACHED [linux/arm/v7 stage-1 13/16] RUN echo "extra-index-url=https://www.piwheels.org/simple" >> .config/pip/pip.conf 0.0s
=> CACHED [linux/arm/v7 stage-1 14/16] RUN pip3 install --upgrade pip && pip3 install --user nltk && pip3 install --user cython && pip3 install --user numpy 0.0s
=> CACHED [linux/arm/v7 stage-1 15/16] RUN if [ `ps auwxf | grep qemu-arm | wc -l` = 4 ]; then pip3 install --user scipy==1.5.0rc2 ; else pip3 install --user scipy ; fi 0.0s
=> [linux/arm/v7 stage-1 16/16] RUN pip3 install --user -r requirements.txt 325.9s
=> [linux/amd64 stage-1 5/16] COPY ./Docker/alpine-cloud/api/entrypoint.sh /home/user_api/entrypoint.sh 1.2s
=> [linux/amd64 stage-1 6/16] RUN touch /home/user_api/skf/db/db.sqlite 0.2s
=> [linux/amd64 stage-1 7/16] RUN chmod a+rw /home/user_api/skf/db /home/user_api/skf/db/* 0.2s
=> [linux/amd64 stage-1 8/16] WORKDIR /home/user_api/ 0.0s
=> [linux/amd64 stage-1 9/16] RUN mkdir .kube 0.2s
=> [linux/amd64 stage-1 10/16] RUN mkdir .config 0.2s
=> [linux/amd64 stage-1 11/16] RUN mkdir .config/pip 0.2s
=> [linux/amd64 stage-1 12/16] RUN echo "[global]" >> .config/pip/pip.conf 0.2s
=> [linux/amd64 stage-1 13/16] RUN echo "extra-index-url=https://www.piwheels.org/simple" >> .config/pip/pip.conf 0.2s
=> [linux/amd64 stage-1 14/16] RUN pip3 install --upgrade pip && pip3 install --user nltk && pip3 install --user cython && pip3 install --user numpy 22.0s
=> [linux/amd64 stage-1 15/16] RUN if [ `ps auwxf | grep qemu-arm | wc -l` = 4 ]; then pip3 install --user scipy==1.5.0rc2 ; else pip3 install --user scipy ; fi 16.0s
=> [linux/amd64 stage-1 16/16] RUN pip3 install --user -r requirements.txt 65.1s
=> ERROR exporting to image 1056.5s
=> => exporting layers 433.2s
=> => exporting manifest sha256:651a8c2a92c21b35939f84379ac3ce6a59a102ad12c1ffd77b87ae5943c10440 0.0s
=> => exporting config sha256:75bc89b31a6e8e5bcb67ef00ed3afbec60af799da0412d17c13852eb956d4e5f 0.0s
=> => exporting manifest sha256:92aad5879eb60c8e4a776ecca27906068d0ae6264b69afefad152e8b566756ad 0.0s
=> => exporting config sha256:24273c4da9572c159210060f87c5889f2f1508feeab36708a27d55fae989945e 0.0s
=> => exporting manifest list sha256:cbbd8a5daf155203ad74613fafddeb7dd1187b7d35da33682c150247c6871c27 0.0s
=> => pushing layers 621.9s
=> => pushing manifest for docker.io/blabla1337/skf-api:dev 1.2s
------
> exporting to image:
------
failed to solve: rpc error: code = Unknown desc = failed commit on ref "manifest-sha256:651a8c2a92c21b35939f84379ac3ce6a59a102ad12c1ffd77b87ae5943c10440": unexpected status: 401 Unauthorized
blabla1337
All 10 comments
So I tried it a couple more times again and now its working:
docker buildx build -f Docker/alpine-cloud/api/Dockerfile --platform linux/amd64,linux/arm/v7 -t blabla1337/skf-api:dev --push .
[+] Building 1148.9s (38/38) FINISHED
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 32B 0.0s
=> [internal] load .dockerignore 0.1s
=> => transferring context: 2B 0.0s
=> [linux/amd64 internal] load metadata for docker.io/library/python:3.7 3.9s
=> [linux/arm/v7 internal] load metadata for docker.io/library/python:3.7 6.2s
=> [internal] load build context 132.2s
=> => transferring context: 967.06MB 122.9s
=> [linux/amd64 stage-1 1/16] FROM docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> => resolve docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> [linux/arm/v7 stage-1 1/16] FROM docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> => resolve docker.io/library/python:3.7@sha256:322ff8fa9fb1cc6d0cc17e2e2b966ca17cda5d036930496cac1a03810525cbb1 0.0s
=> CACHED [linux/amd64 stage-1 2/16] RUN apt update && apt install -y vim libblas-dev libblas-dev liblapack-dev libatlas-base-dev gfortran cython 0.0s
=> CACHED [linux/amd64 stage-1 3/16] RUN groupadd --gid 1000 user_api && useradd --uid 1000 --gid user_api -m user_api && mkdir -p /home/user_api 0.0s
=> CACHED [linux/amd64 stage-1 4/16] ADD ./ /home/user_api/ 0.0s
=> CACHED [linux/amd64 stage-1 5/16] COPY ./Docker/alpine-cloud/api/entrypoint.sh /home/user_api/entrypoint.sh 0.0s
=> CACHED [linux/amd64 stage-1 6/16] RUN touch /home/user_api/skf/db/db.sqlite 0.0s
=> CACHED [linux/amd64 stage-1 7/16] RUN chmod a+rw /home/user_api/skf/db /home/user_api/skf/db/* 0.0s
=> CACHED [linux/amd64 stage-1 8/16] WORKDIR /home/user_api/ 0.0s
=> CACHED [linux/amd64 stage-1 9/16] RUN mkdir .kube 0.0s
=> CACHED [linux/amd64 stage-1 10/16] RUN mkdir .config 0.0s
=> CACHED [linux/amd64 stage-1 11/16] RUN mkdir .config/pip 0.0s
=> CACHED [linux/amd64 stage-1 12/16] RUN echo "[global]" >> .config/pip/pip.conf 0.0s
=> CACHED [linux/arm/v7 stage-1 2/16] RUN apt update && apt install -y vim libblas-dev libblas-dev liblapack-dev libatlas-base-dev gfortran cython 0.0s
=> CACHED [linux/arm/v7 stage-1 3/16] RUN groupadd --gid 1000 user_api && useradd --uid 1000 --gid user_api -m user_api && mkdir -p /home/user_api 0.0s
=> CACHED [linux/arm/v7 stage-1 4/16] ADD ./ /home/user_api/ 0.0s
=> CACHED [linux/arm/v7 stage-1 5/16] COPY ./Docker/alpine-cloud/api/entrypoint.sh /home/user_api/entrypoint.sh 0.0s
=> CACHED [linux/arm/v7 stage-1 6/16] RUN touch /home/user_api/skf/db/db.sqlite 0.0s
=> CACHED [linux/arm/v7 stage-1 7/16] RUN chmod a+rw /home/user_api/skf/db /home/user_api/skf/db/* 0.0s
=> CACHED [linux/arm/v7 stage-1 8/16] WORKDIR /home/user_api/ 0.0s
=> CACHED [linux/arm/v7 stage-1 9/16] RUN mkdir .kube 0.0s
=> CACHED [linux/arm/v7 stage-1 10/16] RUN mkdir .config 0.0s
=> CACHED [linux/arm/v7 stage-1 11/16] RUN mkdir .config/pip 0.0s
=> CACHED [linux/arm/v7 stage-1 12/16] RUN echo "[global]" >> .config/pip/pip.conf 0.0s
=> CACHED [linux/arm/v7 stage-1 13/16] RUN echo "extra-index-url=https://www.piwheels.org/simple" >> .config/pip/pip.conf 0.0s
=> CACHED [linux/arm/v7 stage-1 14/16] RUN pip3 install --upgrade pip && pip3 install --user nltk && pip3 install --user cython && pip3 install --user numpy 0.0s
=> CACHED [linux/arm/v7 stage-1 15/16] RUN if [ `ps auwxf | grep qemu-arm | wc -l` = 4 ]; then pip3 install --user scipy==1.5.0rc2 ; else pip3 install --user scipy ; fi 0.0s
=> [linux/arm/v7 stage-1 16/16] RUN pip3 install --user -r requirements.txt 371.1s
=> [linux/amd64 stage-1 13/16] RUN echo "extra-index-url=https://www.piwheels.org/simple" >> .config/pip/pip.conf 3.7s
=> [linux/amd64 stage-1 14/16] RUN pip3 install --upgrade pip && pip3 install --user nltk && pip3 install --user cython && pip3 install --user numpy 23.3s
=> [linux/amd64 stage-1 15/16] RUN if [ `ps auwxf | grep qemu-arm | wc -l` = 4 ]; then pip3 install --user scipy==1.5.0rc2 ; else pip3 install --user scipy ; fi 12.3s
=> [linux/amd64 stage-1 16/16] RUN pip3 install --user -r requirements.txt 65.4s
=> exporting to image 638.5s
=> => exporting layers 302.5s
=> => exporting manifest sha256:dd1ff1af15e0d11d75c276aa648a56a7debc16bb6242489cb95a9172f4524e7a 0.0s
=> => exporting config sha256:50adb73b1fe9b5e04bf4dd764188f12fabe1a6f693d793a2b06d906edd7dc016 0.0s
=> => exporting manifest sha256:0395b10a6c7045398acb39c4e94e9dc5c3870bf96f001772d0e72b403a523fdb 0.0s
=> => exporting config sha256:15bad4ce8ffeda2ded7e88b7009209fad96589adbde0111a01c35f324601c3f8 0.0s
=> => exporting manifest list sha256:8061b5764cfeeeebf0fc5acb8f0a3bad1475a018a06ca8c0412fbffe56f77ca1 0.0s
=> => pushing layers 332.2s
=> => pushing manifest for docker.io/blabla1337/skf-api:dev 3.6s
No idea what the issue can be...
blabla1337
on 21 Jul 2020
I'm experiencing this in my CI builds, after docker login then a docker buildx and it fails with unauthorized only 1 hour and 1 minute later. How long is this authentication token meant to last? I was sure it would be longer than this.
Here is a full build history if anyone is interested:
https://ci.nginxproxymanager.com/blue/organizations/jenkins/nginx-proxy-manager/detail/master/36/pipeline
EDIT: I should note that retriggering this build works as it uses the cached docker images and pushes faster.
jc21
on 22 Jul 2020
I ran into this problem today creating large images. What I did was use the --cache-to flag when building it, then when I go to push it to dockerhub I log in and use --cache-from.
An example of the workflow would be something like this:
# Build images and cache them to a local cache/ folder, this part takes the longest
docker buildx build --cache-to=type=local,dest=cache,mode=max --platform linux/amd64,linux/arm64 --tag myimagename:latest .
# Login
docker login -u username -p password
# Build again, but use cache from local source
docker buildx build --push --cache-from=type=local,src=cache --platform linux/amd64,linux/arm64 --tag myimagename:latest .
As long as it uploads the image somewhat quickly, it hopefully shouldn't time you out.
Note that this does take a lot more disk space. I made a VM and gave it 64gb of disk space, and with a 7gb docker image crosscompiled to x86 and arm it ran out of disk space while caching all of the layers.
Hope that helps a bit.
EDIT: You could also play around with caching to/from a registry. I mentioned local caching because my internet would be a bottleneck, but it _might_ reduce disk usage if that is a concern, but I havent personally tested it.
emwjacobson
on 30 Jul 2020
Please test with https://github.com/moby/buildkit/pull/1636 docker buildx create --driver-opt image=moby/buildkit:master
tonistiigi
on 25 Aug 2020
congratulation! now I can push. each push is a success, when before it mostly failed (after running for 10 minutes).
thank you very much!
andrew
andrewufrank
on 3 Sep 2020
I'm hitting this at the moment with Quay.io. GitHub Container Registry works OK.
I'm able to reproduce this locally via CLI. The interesting thing that a regular docker push works OK.
Another interesting thing that the first push to Quay.io (when images not yet existed) worked OK.
tyranron
on 24 Nov 2020
@tyranron same deal here: pushing to Quay.io from a GitHub Actions workflow on the subsequent run.
shepmaster
on 4 Dec 2020
I think you guys are hitting https://github.com/containerd/containerd/issues/4622#issuecomment-716746004
tonistiigi
on 4 Dec 2020
Another interesting thing that the first push to Quay.io (when images not yet existed) worked OK.
Exact same issue here with me.
Initial build (fresh repo, no existing tags) succeeds, subsequent builds fail
I've tried with the with.driver-opts: image=moby/buildkit:master in the setup stage to no avail.
Maybe because moby/buildkit#1921 hasn't been approved/merged?
rblaine95
on 3 Jan 2021
Same issue here!
ricardbejarano
on 19 Jan 2021
Related issues
iav
路
7Comments
bjoernhaeuser
路
6Comments
ep4sh
路
5Comments
patrickelectric
路
7Comments
minusdelta
路
10Comments
Most helpful comment
I'm hitting this at the moment with Quay.io. GitHub Container Registry works OK.
I'm able to reproduce this locally via CLI. The interesting thing that a regular
docker pushworks OK.Another interesting thing that the first push to Quay.io (when images not yet existed) worked OK.