DO288 lab2 (ch2s07)

[e719]

hi everybody. I have error when trying to run through lab ch2s07

oc new-app --name elvis \
> quay.io/e719/do288-hello-java
W0220 17:12:43.348779 7576 dockerimagelookup.go:237] container image registry lookup failed: quay.io/e719/do288-hello-java:latest: missing signature key
error: local file access failed with: stat quay.io/e719/do288-hello-java: no such file or directory
error: unable to locate any images in image streams, templates loaded in accessible projects, template files, local docker images with name "quay.io/e719/do288-hello-java"

I make it visible by settings-visibility-make public
A month ago I did other lab from ch2 with do288-apache image. It still there and works fine with oc new-app.

Any ideas?

P.S. Is it special Unix-style challenge to post here a message? While you type in message text, authentication expires, you press post and it shows you ERROR bla-bla-bla, and your message is lost. Now I'm trying to write it again - and it even shows me "autosaved at ... P.M." (not first time it happen)
Very funny

[acshea]

I'm having the same issues today in the same lab! Up until this point, it has never been a problem.

For some reason, using skopeo copy docker-archive with the output of podman save instead of using podman push allowed me to consistently get around the issues.

Other forms of copying that I tried e.g. skopeo copy oci seem to cause the same issues.

[Alexandre]

Hi

No problem today with this lab here using podman push:

1040 podman login -u ran000 quay.io
1041 podman build -t do288-hello-java --layers=false hello-java
1042 podman images
1043 podman tag localhost/do288-hello-java quay.io/ran000/do288-hello-java:latest
1044 podman push quay.io/ran000/do288-hello-java
1045 oc new-app --name elvis quay.io/ran000/do288-hello-java
1046 oc get pods
1047 oc expose svc/elvis
1048 curl elvis-ran000-design-container.apps.ocp4.example.com/api/hello

[acshea]

Issue persists for me today, following the exact steps above. I have deleted my lab environment and started fresh.

Removed login credentials and local instances of the image to prove image was in quay.io and publicly accessible. Full output here:

[student@workstation DO288-apps]$ oc new-app --name elvis quay.io/<user>/do288-hello-java
W0222 08:37:34.982560    6252 dockerimagelookup.go:237] container image registry lookup failed: quay.io/<user>/do288-hello-java:latest: missing signature key
error:  local file access failed with: stat quay.io/<user>/do288-hello-java: no such file or directory
error: unable to locate any images in image streams, templates loaded in accessible projects, template files, local docker images with name "quay.io/<user>/do288-hello-java"

Argument 'quay.io/<user>/do288-hello-java' was classified as an image, image~source, or loaded template reference.

The 'oc new-app' command will match arguments to the following types:

  1. Images tagged into image streams in the current project or the 'openshift' project
     - if you don't specify a tag, we'll add ':latest'
  2. Images in the Docker Hub, on remote registries, or on the local Docker engine
  3. Templates in the current project or the 'openshift' project
  4. Git repository URLs or local paths that point to Git repositories

--allow-missing-images can be used to point to an image that does not exist yet.

See 'oc new-app -h' for examples.
[student@workstation DO288-apps]$ podman rmi -af
........
........
[student@workstation DO288-apps]$ podman logout quay.io
Removed login credentials for quay.io
[student@workstation DO288-apps]$ podman pull quay.io/<user>/do288-hello-java
Trying to pull quay.io/<user>/do288-hello-java:latest...
........
........
[student@workstation DO288-apps]$ podman images
REPOSITORY                       TAG     IMAGE ID      CREATED        SIZE
quay.io/<user>/do288-hello-java  latest  edebfff41b59  5 minutes ago  459 MB

Interestingly, using the --allow-missing-images flag allows for the application pod to be created and pulls the image successfully, however certain resources are not created such as the service. 

[mohamedarif]

I am also facing the same issue any help will be appreciated. 

[mohamedarif]

If anyone else facing this issue. Here is workaround and tracking for your information 

As a workaround kindly try the below suggestion.
Add the `--format v2s1` flag to the skopeo copy command in the previous step to push the image to Quay with an OCP4.6 compatible manifest.

Refer to https://issues.redhat.com/browse/PROJQUAY-3285 to track the long term fix.

Alternatively, you can build the image by using the podman build --format=docker in step 3.3.

podman build --format=docker --layers=false -t do288-apache ./container-build

[DerekLyons]

This fix worked for me! Just added --format=docker like he said above.

[rrgoncalves]

This fix worked for me! 

[gamagnolo]

Same for DO288 ch9s2.

It doesn't seem resolved...

Can anyone confirm ?

Thanks

[mohamedarif]

It did resolved the issue. During podman build I added --format v2s1

Without that I was always getting that error. This is the status 1 week before. Didnt check recently.