build-push-action
build-push-action copied to clipboard
error: failed to solve: failed to compute cache key: failed to copy: invalid status response 416
Behaviour
Got an error while computing the cache key:
error: failed to solve: failed to compute cache key: failed to copy: invalid status response 416 The range specified is invalid for the current size of the resource. for https://vth0acprodeus2file0.blob.core.windows.net/1665cc709f9c430f88b363d79c2f9f6f/35da92eddf84ec1194f60050f276280d?sv=2019-07-07&sr=b&sig=tAOoxQlbhCkNGNswTquFfnmtVgq3VCAP4Icc0bdnvlg%3D&se=2022-02-08T15%3A07%3A22Z&sp=r&rscl=x-e2eid-d10f7757-becb4d8c-94279385-0bd5086d
[4181](https://github.com/dfinity/internet-identity/runs/5110353933?check_suite_focus=true#step:4:4181)
Error: buildx failed with: error: failed to solve: failed to compute cache key: failed to copy: invalid status response 416 The range specified is invalid for the current size of the resource. for https://vth0acprodeus2file0.blob.core.windows.net/1665cc709f9c430f88b363d79c2f9f6f/35da92eddf84ec1194f60050f276280d?sv=2019-07-07&sr=b&sig=tAOoxQlbhCkNGNswTquFfnmtVgq3VCAP4Icc0bdnvlg%3D&se=2022-02-08T15%3A07%3A22Z&sp=r&rscl=x-e2eid-d10f7757-becb4d8c-94279385-0bd5086d
This was transient and I'm not sure how to reproduce it. Here is the GitHub actions run: https://github.com/dfinity/internet-identity/runs/5110353933?check_suite_focus=true
I reckon that might be a buildx
issue more than a build-push-action
issue; happy to migrate the ticket if you confirm.
Getting this somewhat regularly now, any idea what might be happening? https://github.com/dfinity/internet-identity/runs/5156407802?check_suite_focus=true#step:4:2065
@nmattia Looks linked to the GitHub Cache API. @dhadka @chrispat We saw that one too few days ago with @tonistiigi. Any ideas?
It's gotten less frequent, but still running into it every now and then: https://github.com/dfinity/internet-identity/runs/5553785037?check_suite_focus=true
Got a similar problem too: https://github.com/litentry/litentry-parachain/runs/5576261715?check_suite_focus=true
I’m also facing this issue across multiple projects 😞 this is the setup we’re using
- uses: docker/build-push-action@v2
with:
context: .
push: true
tags: ghcr.io/org/repo:tag
cache-from: type=gha
cache-to: type=gha,mode=max
We updated the error message for this case in v0.10.1 with extra info. If you have received this error after that then please post the message here.
@tonistiigi to clarify, this is v0.10.1
of what?
@nmattia BuildKit 0.10.1 which is the latest stable so it should be the one used in your workflow if you have not enforced a specific BuildKit image in setup-buildx-action
.
@dhadka
Got this on buildkit CI.
error: failed to solve: failed to compute cache key: failed to copy: invalid status response 416 The range specified is invalid for the current size of the resource. for https://ki6cacprodeus1file5.blob.core.windows.net/2ff2ce0399e5464697755b110c91788d/dd15048612dcec11b65614cb655b2cad?sv=2019-07-07&sr=b&sig=n7RwHrvaJu%2FrMLyCzY3e%2FXnNBEImbs0XulVQstSARV8%3D&se=2022-05-26T01%3A48%3A36Z&sp=r&rscl=x-e2eid-e0b78c9e-7a194019-bf7b385e-d8b4c39f, range: bytes=115194599-
But the URL is already expired so I can't inspect what is wrong with that blob. Maybe you have better visibility to give some ideas. I wonder if bug could be that this is asking for the range exactly at the end of the file.
Now I am unable to use either ADD or COPY in my Dockerfile so the build process fails. I am receiving the errors listed below:
ERROR: failed to calculate checksum of ref 1gjng5kin5uk6yv60uy8tx1hs::w6ujc2j2e2mnienr66n7godbb: failed to walk /tmp/buildkit-mount296995550/tmp: lstat /tmp/buildkit-mount296995550/tmp: no such file or directory
error: failed to solve: failed to compute cache key: failed to calculate checksum of ref 1gjng5kin5uk6yv60uy8tx1hs::w6ujc2j2e2mnienr66n7godbb: failed to walk /tmp/buildkit-mount296995550/tmp: lstat /tmp/buildkit-mount296995550/tmp: no such file or directory
Error: buildx failed with: error: failed to solve: failed to compute cache key: failed to calculate checksum of ref 1gjng5kin5uk6yv60uy8tx1hs::w6ujc2j2e2mnienr66n7godbb: failed to walk /tmp/buildkit-mount296995550/tmp: lstat /tmp/buildkit-mount296995550/tmp: no such file or directory
I used Dockerfile file in the home folder or under another folder, nothing changed.
same
root@VM-20-2-ubuntu:~/test# cat Dockerfile
FROM quay.io/skopeo/stable:latest as builder1
RUN for image in $(cat images.list); do skopeo sync --format=v2s2 --src docker --dest docker ${image} localhost:5000 --dest-tls-verify=false; do
ne
FROM registry:2
COPY --from=builder1 /tmp /tmp
COPY registry /var/lib/registry/
root@VM-20-2-ubuntu:~/test# ll
total 12
drwxr-xr-x 2 root root 4096 Jun 14 16:52 ./
drwx------ 12 root root 4096 Jun 14 16:52 ../
-rw-r--r-- 1 root root 603 Jun 14 16:52 Dockerfile
build failed, must mkdir test/registry
before build , but the test/registry
will generate auto after build !!
root@VM-20-2-ubuntu:~/test# docker buildx build -t test .
[+] Building 1.4s (14/15)
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 640B 0.0s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [internal] load metadata for docker.io/library/registry:2 1.0s
=> [internal] load metadata for quay.io/skopeo/stable:latest 0.3s
=> [internal] load metadata for quay.io/kubespray/kubespray:v2.19.0 0.0s
=> [builder0 1/3] FROM quay.io/kubespray/kubespray:v2.19.0 0.0s
=> [internal] load build context 0.0s
=> => transferring context: 2B 0.0s
=> CACHED [stage-2 1/3] FROM docker.io/library/registry:2@sha256:bedef0f1d248508fe0a16d2cacea1d2e68e899b2220e2258f1b604e1f327d475 0.4s
=> [builder1 1/3] FROM quay.io/skopeo/stable:latest@sha256:4dd7606ce9098541b10b99920da3bfd9740e004b9862f9822ed7d8d0bf6ea870 0.0s
=> CACHED [builder0 2/3] RUN cd /kubespray && grep -rIl . | xargs sed -i "s/\r//g" || true 0.0s
=> CACHED [builder0 3/3] RUN cd /kubespray/contrib/offline/ && bash generate_list.sh 0.0s
=> CACHED [builder1 2/3] COPY --from=builder0 /kubespray/contrib/offline/temp/images.list . 0.0s
=> CANCELED [builder1 3/3] RUN for image in $(cat images.list); do skopeo sync --format=v2s2 --src docker --dest docker ${image} localho 0.4s
=> CACHED [stage-2 2/3] COPY --from=builder1 /tmp /tmp 0.0s
=> ERROR [stage-2 3/3] COPY registry /var/lib/registry/ 0.0s
------
> [stage-2 3/3] COPY registry /var/lib/registry/:
------
error: failed to solve: failed to compute cache key: "/registry" not found: not found
root@VM-20-2-ubuntu:~/test#
Also running into this, both locally and in github actions
I got the error as well. Any idea how to solve the error?
Never mind. I figured it out. My COPY command should be
COPY ./project/techtrends/ /app
instead of
COPY ./techtrends/ /app
because techtrends is under project directory
I've never had issues with this COPY line before but it's happening consistently when building with load: true
in GH actions.
Dockerfile:6
--------------------
4 |
5 | COPY --from=pandoc /usr/local/bin/pandoc* /usr/bin/
6 | >>> COPY --from=pandoc /opt/texlive* /usr/bin/
7 |
8 | ARG BUILD_PACKAGES="build-essential yarn git-core ..."
--------------------
error: failed to solve: failed to compute cache key: failed to calculate checksum of ref xvy7rzjibaifonylswzxq9vel::ma4wp1egg53q7n0wq3kbi98qf: "/opt/texlive/texdir/texmf-dist/doc/man": not found
Error: buildx failed with: error: failed to solve: failed to compute cache key: failed to calculate checksum of ref xvy7rzjibaifonylswzxq9vel::ma4wp1egg53q7n0wq3kbi98qf: "/opt/texlive/texdir/texmf-dist/doc/man": not found
Update:
Seems this was just because COPY --from=pandoc /opt/texlive* /usr/bin/
does not maintain folder structure. Changing it to COPY --from=pandoc /opt/texlive/ /usr/bin/texlive/
fixed it. Details here: https://stackoverflow.com/questions/30215830/dockerfile-copy-keep-subdirectory-structure
I've got the current problem that the COPY --from=builder ./app/stackable-kafka-operator/ ./stackable-kafka-operator
where FROM docker.stackable.tech/stackable-experimental/ubi8-rust-builder AS builder
This error happened the first time, when switching from build to buildx to enable multi-arch builds (which is quite painful anyway with qemu).
I've tried several things to avoid that error:
-
COPY --from=builder ./app/stackable-kafka-operator/ ./
-
COPY --from=builder ./app/stackable-kafka-operator/ /
-
COPY --from=builder /app/stackable-kafka-operator/ /
-
COPY --from=builder ./app/stackable-kafka-operator /
-
COPY --from=builder /app/stackable-kafka-operator /
Sadly none of those combination was working.
To compile the operator during building time of the image we use a ubi8-minimal based rust builder. Multi-Arch is tried to be archived leveraging cargo build --target=ARCH
(played some tricks here to archive building under qemu, testing of the new things fails because the arm64 side of things (native) fails during build time at this line since it is way faster.
Do I overlook something or is this an actual problem?
It fails with:
error: failed to solve: failed to compute cache key: failed to calculate checksum of ref rykfgneaumnifqjl5s76cl050::fvg8fineqh20z2i8y7dx5t3a2: "/app/stackable-kafka-operator": not found
Nevermind: COPY ./app/stackable-kafka-operator* ./stackable-kafka-operator
worked for me
I'm getting this same issue consistently, even with the latest buildkit:
moby/buildkit:buildx-stable-1 => buildkitd github.com/moby/buildkit v0.10.4 a2ba6869363812a210fcc3ded6926757ab780b5f
@Maleware There is no such path /app/stackable-kafka-operator
anywhere in the docker.stackable.tech/stackable-experimental/ubi8-rust-builder
image afaics.
I am also facing this docker build command's issue. When I run: docker build -t myimage .
The following is what I got:
> [2/5] COPY . /app:
------
failed to compute cache key: failed to copy: httpReadSeeker: failed open: failed to do request: Get "https://d2glxqk2uabbnd.cloudfront.net/b16f79-305812972329-ecbbcc52-aa8e-da32-9e5b-1acde3109751/b8500d72-cb2f-4ffe-8e1d-314cfa1121af?Expires=1665868609&Signature=SfNnQoCyOEjU1~p9hpfd~6lMml1le-9SQWMv0xv9X53eCHjC9C~qFdlKY9ir7PQG7H2zYFQsv7uju61-9YOEFaT2BDTSVoBPxmrypZI-VRmZzVkGe5~nZG~~eFRIt3dcIXHyVChpVd1WcsEQqe4Mem08vS7hjVZoKqKJEqT7EVwqgGDn~rdoTuKRYpdK8lKacgS3HALiEIH3Tv~PU4c2HB61eoVqIBCVOnDmQCzCdwje-hhtUja5r8O2L2tr6ns1proFFFe5uqqQUai9Z3QZIRlL-~JRDKjZtg~eZCyJL7WBoD1~u7meZMgEOjR4XrQ0pZyvgRo7cY1lxpV2dZTnlg__&Key-Pair-Id=KSPLJWGOARK62": net/http: TLS handshake timeout
Here is my Dockerfile
FROM public.ecr.aws/sam/build-python3.7:latest
COPY . /app
WORKDIR /app
RUN pip install --upgrade pip
RUN pip install -r requirements.txt
ENTRYPOINT ["gunicorn", "-b", ":8080", "main:APP"]
~~Same issue here, building works fine locally.~~
~~Failed build: https://github.com/Shoutz0r/frontend/actions/runs/3308523917/jobs/5460967584~~ ~~Dockerfile: https://github.com/Shoutz0r/frontend/blob/4d2d50968d0a222f512f4c16660547c7f6df0b78/Dockerfile~~
EDIT:
Turns out that the docker/build-push-action@v3
uses the git
context by default for building the docker images, rather then the local directory. Had to set context: .
to fix this. Documentation
Oh, thank you. I have resolved the issue.
On Sun, Oct 23, 2022 at 4:56 PM Jorin Vermeulen @.***> wrote:
Same issue here, building works fine locally.
Failed build: https://github.com/Shoutz0r/frontend/actions/runs/3308523917/jobs/5460967584 Dockerfile: https://github.com/Shoutz0r/frontend/blob/4d2d50968d0a222f512f4c16660547c7f6df0b78/Dockerfile
— Reply to this email directly, view it on GitHub https://github.com/docker/build-push-action/issues/558#issuecomment-1288199572, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHSVXERW3USIQK6RRH3CE2LWEWRBLANCNFSM5N2TSRZA . You are receiving this because you commented.Message ID: @.***>
Closing this issue as many comments are not related to the initial report.
@nmattia If you still encounter this issue, can you provide another repro with also BuildKit logs? Thanks.
@crazy-max will do thanks!
I set the context
,but I still have the error.
https://github.com/frank-pian/OCCT/blob/occ-api/.github/workflows/build.yml
Dockerfile:51
--------------------
49 | RUN git clone --depth 1 https://github.com/gkv311/tcl.git
50 | WORKDIR tcl
51 | >>> ADD occ_tcl_build_wasm.sh .
52 | RUN /bin/bash occ_tcl_build_wasm.sh
53 |
--------------------
ERROR: failed to solve: failed to compute cache key: failed to calculate checksum of ref lvqspldaya570v6mf2cna5bxq::up36el7zdf0pt6k9d9cmrl9tx: "/occ_tcl_build_wasm.sh": not found
Error: buildx failed with: ERROR: failed to solve: failed to compute cache key: failed to calculate checksum of ref lvqspldaya570v6mf2cna5bxq::up36el7zdf0pt6k9d9cmrl9tx: "/occ_tcl_build_wasm.sh": not found
I just got this error tonight. If anyone stumbles across this in the future, adding DOCKER_BUILDKIT=0
before the build command worked for me.
We are getting this error --- happens every once in a while in our GH Actions for whatever reason:
#22 [16/17] COPY . /app
#22 ERROR: blob not found
------
> [16/17] COPY . /app:
------
Dockerfile:122
--------------------
12*** | fi
121 |
122 | >>> COPY . /app
123 | RUN mkdir -p /app/temp_files
124 | EXPOSE 8*** 8***82
--------------------
error: failed to solve: failed to compute cache key: blob not found
Error: buildx failed with: error: failed to solve: failed to compute cache key: blob not found
We have not been able to solve it and our builds are stalled.
Same here.
https://github.com/theAkito/docker-murmur/actions/runs/5627573725/job/15250407075#step:7:6824
What's the actual reason behind this issue?
@theAkito
#24 [linux/amd64 stage-1 5/6] COPY --from=build /root/mumble/release/murmurd /usr/bin/murmurd
#24 ERROR: failed to calculate checksum of ref qgp9ucmtmblbgry83pja48n5y::8myxzhj2gtzqfcqjue0dzjw4i: failed to walk /tmp/buildkit-mount2545155800/root/mumble/release: lstat /tmp/buildkit-mount2545155800/root/mumble/release: no such file or directory
This does not look related, can you open a new issue please?