Description of problem: When trying to push an Ansible Playbook Bundle (Container) from a development workstation to an Ansible Service Broker it discovers the Docker Registry to push to. However it discovers the Service IP (172.30....) instead of the route. Consequently it is not possible to push a built APB Container from a workstation that is not part of the OpenShift SDN. This should be changed to use the Registry route rather than the Service. Version-Release number of selected component (if applicable): Latest in the 3.7 OCP repos (yum -y install apb) How reproducible: Steps to Reproduce: From a RHEL Machine (with OCP Subscription): 1. yum -y install apb 2. git clone https://github.com/ansibleplaybookbundle/etherpad-apb 3. cd etherpad-apb 4. apb build 5. apb push --openshift --broker https://asb-1338-openshift-ansible-service-broker.apps.rdu.example.opentlc.com/ansible-service-broker (replacing with the correct ASB Route) Actual results (only output from apb push is shown): ocplab-07ef ~/etherpad-apb (master) $ apb push --openshift --broker https://asb-1338-openshift-ansible-service-broker.apps.rdu.example.opentlc.com/ansible-service-broker version: 1.0 name: etherpad-apb description: Note taking web application bindable: True async: optional metadata: documentationUrl: https://github.com/ether/etherpad-lite/wiki imageUrl: https://translatewiki.net/images/thumb/6/6f/Etherpad_lite.svg/200px-Etherpad_lite.svg.png dependencies: ['docker.io/mariadb:latest', 'docker.io/tvelocity/etherpad-lite:latest'] displayName: Etherpad (APB) longDescription: An apb that deploys Etherpad Lite providerDisplayName: "Red Hat, Inc." plans: - name: default description: A single etherpad application with no DB free: true metadata: displayName: Default longDescription: This plan provides a single Etherpad application with no database cost: $0.00 parameters: - name: mariadb_name required: true default: etherpad type: string title: MariaDB Database Name - name: mariadb_user required: true default: etherpad title: MariaDB User type: string maxlength: 63 - name: mariadb_password default: admin type: string description: A random alphanumeric string if left blank title: MariaDB Password - name: mariadb_root_password default: admin type: string description: root password for mariadb title: Root Password /usr/lib/python2.7/site-packages/urllib3/connectionpool.py:852: InsecureRequestWarning: Unverified HTTPS request is being made. Adding certificate verification is strongly advised. See: https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings InsecureRequestWarning) Found registry IP at: 172.30.27.190:5000 Building image with the tag: 172.30.27.190:5000/openshift/etherpad-apb Error accessing the docker API. Is the daemon running? Exception occurred! 500 Server Error: Internal Server Error ("Get https://172.30.27.190:5000/v1/users/: dial tcp 172.30.27.190:5000: i/o timeout") Expected results: [...] Found registry at: https://docker-registry-default.apps.rdu.example.opentlc.com Additional info: This same sequence of commands works if executed on the master. Which of course is unacceptable - Developers will never get access to the master.
This issue should duplicate with https://bugzilla.redhat.com/show_bug.cgi?id=1519193
Dylan please help answer comment #3
Hey Gajanan, There is a workaround for this and I am in the progress of submitting some patches upstream to make this process easier. We have a similar bug filed here: https://bugzilla.redhat.com/show_bug.cgi?id=1519193. To workaround this the customer can take note of the registry route: $ oc get route docker-registry -n default docker_registry.default.<cluster_ip>.nip.io Then build the container image with apb build specifying the proper route: $ apb build --tag <registry_route>/openshift/<my-apb> $ docker login <registry_route> -u unused -p $(oc whoami -t) $ docker push <registry_route>/openshift/<my-apb> $ apb bootstrap This will push the image to the local OpenShift Container Registry and then force the OpenShift Ansible Broker to bootstrap and relist the Service Catalog.
Good start for a workaround. However it fails on docker login since my cluster has self signed certificates - which a lot of development clusters will have. docker login docker-registry-default.apps.wk.example.opentlc.com -u unused -p $(oc whoami -t) Error response from daemon: Get https://docker-registry-default.apps.wk.example.opentlc.com/v1/users/: x509: certificate signed by unknown authority
Wolfgang, Is it out of the question for the user to add the docker-registry route to the list of insecured registries in /etc/sysconfig/docker? This is really an issue with the Docker client than it is the APB tooling. The tooling (even if we weren't using the workaround) expects all of the configuration for the registry to be done in the Docker config files. I tested adding the route to the insecure registry list (--insecure-registry docker-registry-default.apps.wk.example.opentlc.com in /etc/sysconfig/docker for you) and restarting Docker solved this problem. Please let me know if that approach doesn't work for you. Unfortunately its our only workaround since we are dependent on the Docker client.
Dylan, we are getting somewhere. :-) In the current Docker 1.12.6 the location is /etc/containers/registries.conf though instead of /etc/sysconfig/docker. Regardless that worked. :-) Now to push to ..../openshift/... the user OpenShift user that tries the push seems to need cluster-admin authorization. Also the example above needs the full broker URL to succeed: apb bootstrap --broker https://asb-1338-openshift-ansible-service-broker.apps.wk.example.opentlc.com/ansible-service-broker Now finally when while the push and bootstrap succeeds a "apb list" does not show the new APB (I should have a mariadb-apb (not the rh- one) and a rocketchat-apb). apb list --broker https://asb-1338-openshift-ansible-service-broker.apps.wk.example.opentlc.com/ansible-service-broker ID NAME DESCRIPTION 2c259ddd8059b9bc65081e07bf20058f rh-mariadb-apb Mariadb apb implementation 03b69500305d9859bb9440d9f9023784 rh-mediawiki-apb Mediawiki123 apb implementation 73ead67495322cc462794387fa9884f5 rh-mysql-apb Software Collections MySQL APB d5915e05b253df421efe6e41fb6a66ba rh-postgresql-apb SCL PostgreSQL apb implementation I feel this is something different alltogether though.
Wolfgang, Yes it is true that the user you are using to login to the registry needs to have cluster-admin permissions. The APB tooling is heavily geared towards a user with cluster-admin access. I also submitted a PR to this bug https://bugzilla.redhat.com/show_bug.cgi?id=1523252 which will allow you to do: apb bootstrap --broker https://asb-1338-openshift-ansible-service-broker.apps.wk.example.opentlc.com instead of the full route with suffix. I am willing to bet this is an issue with the local_openshfit registry whitelist. By default a cluster has an empty whitelist for this registry adapter. Please go here: https://github.com/openshift/ansible-service-broker/blob/master/docs/config.md#local-openshift-registry and note the `white_list` config value. As a user with access to the ansible-service-broker namespace (or openshift-ansible-service-broker using openshift-ansible) you can do: oc edit configmap broker-config -n ansible-service-broker and add the whitelist value so that all images that end in `-apb` will be bootstrapped. This should resolve your problem.
Also, I added a PR to openshift-docs https://github.com/openshift/openshift-docs/pull/6780 forever ago to add these instructions into the openshift-ansible installer docs. Hopefully the info in there helps.
Awesome! Got it running! I did add the whitelist to the local_openshift stanza. And sure enough bootstrap/list (and even deploy of the RocketChat apb) worked like a charm. Question: rather than using just 'openshift' under 'namespaces' for local_openshift is it possible to use '*'? Or a setting that would scan every OpenShift project? For my own testing I could certainly add a project or two - but for a "real" developer environment I feel that might be a bit tedious. Thanks for all your help getting this running!!
Great! Unfortunately we don't currently have that ability but it seems trivial to add it. I created a GitHub issue here: https://github.com/openshift/ansible-service-broker/issues/623. We default to the openshift namespace because it is unique in that is makes all of its images/imagestreams available to the entire cluster. Creating a new project by default will not have its images be globally accessible. Regardless you bring up a valid use case so we can track that functionality in the open issue.
https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/pull/187 The above PR adds the ability to specify a registry route so that apb push can be used outside the SDN.
Moving to ON_QA. The important thing to test is being able to specify --registry-route when pushing.
Jian, --registry-route is used if you do not have access to the `default` namespace where the OpenShift registry lives. This way you can push to the internal registry without reading the IP address of the registry service. You are using it properly but it looks like your Docker daemon is not running on your host. You need Docker installed where the APB tool is in order to use the docker client to login to the internal registry. Please start the Docker daemon and try again.
Dylan, Sorry, I forgot to post my Docker running info. My docker daemon was running well. Like below: [root@localhost hello-world-apb]# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES [root@localhost hello-world-apb]# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES [root@localhost hello-world-apb]# docker images REPOSITORY TAG IMAGE ID CREATED SIZE docker.io/ansibleplaybookbundle/apb-tools sprint143.1 79b05404db24 3 days ago 730.2 MB docker.io/zjianbjz/hello-world-db-apb latest 3ad4a49a0bf9 6 days ago 672.4 MB
#FYI [root@localhost hello-world-apb]# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES [root@localhost hello-world-apb]# ps -elf | grep docker 4 S root 12730 1 0 80 0 - 124204 futex_ Jan22 ? 00:00:02 /usr/libexec/docker/docker-containerd-current --listen unix:///run/containerd.sock --shim /usr/libexec/docker/docker-containerd-shim-current --start-timeout 2m 4 S root 29579 1 1 80 0 - 145609 futex_ 09:37 ? 00:00:00 /usr/bin/dockerd-current --add-runtime oci=/usr/libexec/docker/docker-runc-current --default-runtime=oci --containerd /run/containerd.sock --exec-opt native.cgroupdriver=systemd --userland-proxy-path=/usr/libexec/docker/docker-proxy-current --selinux-enabled --log-driver=journald --insecure-registry 172.30.0.0/16 --insecure-registry asb-registry.usersys.redhat.com:5000 0 S root 29723 24258 0 80 0 - 29843 pipe_w 09:37 pts/1 00:00:00 grep --color=auto docker
Jian, Ah okay, I actually think the problem here is that you are using an https route. The --registry-route flag prefixes the image tag with the --registry-route and Docker does not accept http or https in the tag. Hence: Exception occurred! 500 Server Error: Internal Server Error ("Error parsing reference: "https://docker-registry-default.apps.0122-0tk.qe.rhcloud.com/openshift/hello-world-apb" is not a valid repository/tag") If your registry is secured with TLS (which by default in openshift its a self signed cert) then you must be sure to update your docker configuration such that docker-registry-default.apps.0122-0tk.qe.rhcloud.com is added as an insecure registry. These settings are in /etc/sysconfig/docker and you want to add the flag `--insecure-registry docker-registry-default.apps.0122-0tk.qe.rhcloud.com`. Hope that helps. Please retest without https on the tag.
Jian, Please reread comment #22. The certificate warning is because you need to add the docker registry route as an insecure registry in the Docker config. We are currently tracking a way to avoid this step by using the native Docker client on the remote host, but the work is not done yet. Can you please retry with the proper Docker config?
Jian, Thanks for retesting! This looks more like the bug I would expect to see if the login is failing.
I have changed this bug to describe the timeout errors seen when using `apb push` against a cluster that does not exist on the same host where the tool is run. This is seen in a Minishift environment as well. The proper way to solve this would be to connect to the remote host's docker daemon from the APB tooling the same way that Minishift is supported now. This will be difficult because we need to get the certs from the remote host. I am going to use this bug to track what changes are necessary to improve this process for a developer.
*** Bug 1519193 has been marked as a duplicate of this bug. ***
We have documented a workaround for working with remote clusters here: https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/blob/master/docs/developers.md#alternative-to-using-apb-push Instead of using `apb push` the developer can follow this documentation approach to populate their image onto the OpenShift cluster.
Changing component to Documentation since we have properly documented the workaround here: https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/blob/master/docs/developers.md#alternative-to-using-apb-push. We need this documented for 3.10 release showing how to populate internal openshift registry with APB images using `oc new-app`.
*** Bug 1537599 has been marked as a duplicate of this bug. ***
*** Bug 1541903 has been marked as a duplicate of this bug. ***
Dylan, I followed your workaround, but got nothing when running `oc get images| grep <bundle_name>`, I think we should change it to `oc get imagestreams -n openshift | grep <bundle_name>`, details: 1) #oc new-app <path_to_bundle_source> --name <bundle_name> -n openshift [jzhang@localhost hello-world-apb]$ oc new-app . --name jian -n openshift --> Found Docker image f29c340 (15 hours old) from Docker Hub for "ansibleplaybookbundle/apb-base" * An image stream will be created as "apb-base:latest" that will track the source image * A Docker build using source code from git:ansibleplaybookbundle/hello-world-apb.git#master will be created * The resulting image will be pushed to image stream "jian:latest" * Every time "apb-base:latest" changes a new build will be triggered * WARNING: this source repository may require credentials. Create a secret with your git credentials and use 'set build-secret' to assign it to the build config. * This image will be deployed in deployment config "jian" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/jian --port=[port]' later * WARNING: Image "ansibleplaybookbundle/apb-base" runs as the 'root' user which may not be permitted by your cluster administrator --> Creating resources ... imagestream "apb-base" created imagestream "jian" created buildconfig "jian" created deploymentconfig "jian" created --> Success Build scheduled, use 'oc logs -f bc/jian' to track its progress. Run 'oc status' to view your app. 2)Got nothing: [root@host-172-16-120-36 ~]# oc get images| grep jian [root@host-172-16-120-36 ~]# oc get imagestreams -n openshift| grep jian jian docker-registry.default.svc:5000/openshift/jian
Jian, If the imagestream was created that means that the image was more than likely in the process of building. You'll see the last line is "Build scheduled". You can watch the logs of the build and when it's completed the image will be pushed to the internal registry which at that point `oc get images` will show the APB.
This bug is being updated to reflect a documentation update for 3.10 that will explain the limitation of pushing images to a registry on a remote host. To workaround this, a developer can use `oc new-app` as documented here: https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/blob/master/docs/developers.md#alternative-to-using-apb-push. For `apb run` and `apb test`, these commands directly depend upon `apb push` working properly. This means that if a user wants to do something similar to `apb run` they should instead use `oc new-app` to get the image onto the internal registry, then use `oc run` to manually deploy the pod. `apb test` is really no different from `apb run` except that it runs the specific action `test` so it has the same limitations as `apb run`.
This bug needs to include instructions on how to `oc run` an APB to workaround `apb run` not working. I will include a document upstream and link it here.
https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/pull/278
Dylan, I followed your workaround, I think ` cat apb.yml | grep base64` should change to `$cat apb.yml | base64`.
sunzhaohua, Thanks for catching that! I submitted a fix along with this PR: https://github.com/ansibleplaybookbundle/ansible-playbook-bundle/pull/281
Dylan, 1) I met the same problem with jian at comment 35. [szh@localhost my-doc-apb]$ oc new-app ./ --name my-doc -n openshift --> Found Docker image 57595a3 (2 weeks old) from Docker Hub for "ansibleplaybookbundle/apb-base" * An image stream will be created as "apb-base:latest" that will track the source image * A Docker build using binary input will be created * The resulting image will be pushed to image stream "my-doc:latest" * A binary build was created, use 'start-build --from-dir' to trigger a new build * This image will be deployed in deployment config "my-doc" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/my-doc --port=[port]' later * WARNING: Image "ansibleplaybookbundle/apb-base" runs as the 'root' user which may not be permitted by your cluster administrator --> Creating resources ... imagestream "apb-base" created imagestream "my-doc" created buildconfig "my-doc" created deploymentconfig "my-doc" created --> Success Build scheduled, use 'oc logs -f bc/my-doc' to track its progress. Run 'oc status' to view your app. [szh@localhost my-doc-apb]$ oc logs -f bc/my-doc -n openshift error: no builds found for "my-doc" [szh@localhost my-doc-apb]$ curl -H "Authorization: Bearer $(oc whoami -t)" -k https://asb-1338-openshift-ansible-service-broker.apps.0528-l3l.qe.rhcloud.com/ansible-service-broker/v2/catalog | grep my-doc % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 3713 0 3713 0 0 3306 0 --:--:-- 0:00:01 --:--:-- 3309 2) In step3, "oc get route -n ansible-service-broker" should be changed to "oc get route -n openshift-ansible-service-broker"
Yes for step 2 the namespace should be `openshift-ansible-service-broker`. This change should be in the docs that Alex submits, but upstream our namespace is `ansible-service-broker`. I also do not see any errors in the `oc new-app` command which tells me that a buildconfig was created. I am not sure why you are seeing "no build found" on the next call. Can you give me more information about what is created after that call? You should see imagestreams for `apb-base` and `my-doc`. Can you return the output of the following: $ oc get is --all-namespaces | grep my-doc $ oc get bc --all-namespaces | grep my-doc $ oc get images | grep apb-base $ oc get images | grep my-doc It's also possible that on your cluster the user is not allowed to create buildconfigs in namespace `openshift` but I did not have this problem on my cluster. Any more info you can give me would be helpful because the actual `new-app` command succeeded and on my cluster it creates an associated buildconfig.
https://github.com/openshift/openshift-docs/pull/9648
Comment 35 Comment 43 verified successful use hello-world-apb (https://github.com/ansibleplaybookbundle/hello-world-apb) failed use my own apb 1) use hello-world-apb [szh@localhost hello-world-apb]$ oc new-app ./ --name hellooooo -n openshift --> Found Docker image 486a293 (7 days old) from Docker Hub for "ansibleplaybookbundle/apb-base:canary" * An image stream will be created as "apb-base:canary" that will track the source image * A Docker build using source code from https://github.com/ansibleplaybookbundle/hello-world-apb.git#master will be created * The resulting image will be pushed to image stream "hellooooo:latest" * Every time "apb-base:canary" changes a new build will be triggered * This image will be deployed in deployment config "hellooooo" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/hellooooo --port=[port]' later * WARNING: Image "ansibleplaybookbundle/apb-base:canary" runs as the 'root' user which may not be permitted by your cluster administrator --> Creating resources ... imagestream "hellooooo" created buildconfig "hellooooo" created deploymentconfig "hellooooo" created --> Success Build scheduled, use 'oc logs -f bc/hellooooo' to track its progress. Run 'oc status' to view your app. [szh@localhost hello-world-apb]$ oc logs -f bc/hellooooo -n openshift Cloning "https://github.com/ansibleplaybookbundle/hello-world-apb.git" ... Commit: 7132053ccc8ea64b856074247638cfde54feaaff (Skip last operation if not in cluster (#6)) Author: David Zager <dzager> Date: Thu Apr 12 12:06:42 2018 -0400 Pulling image ansibleplaybookbundle/apb-base@sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b ... Pulled 0/14 layers, 0% complete Pulled 1/14 layers, 7% complete Pulled 2/14 layers, 16% complete Pulled 3/14 layers, 25% complete Pulled 4/14 layers, 34% complete Pulled 5/14 layers, 41% complete Pulled 6/14 layers, 49% complete Pulled 7/14 layers, 57% complete Pulled 8/14 layers, 59% complete Pulled 9/14 layers, 66% complete Pulled 10/14 layers, 73% complete Pulled 11/14 layers, 80% complete Pulled 12/14 layers, 88% complete Pulled 13/14 layers, 95% complete Pulled 14/14 layers, 100% complete Extracting Step 1/9 : FROM ansibleplaybookbundle/apb-base@sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b ---> 486a2936c6a8 Step 2/9 : ENV "HTTP_PROXY" "http://file.rdu.redhat.com:3128" "HTTPS_PROXY" "http://file.rdu.redhat.com:3128" "NO_PROXY" ".centralci.eng.rdu2.redhat.com,.cluster.local,.svc,169.254.169.254,172.16.120.62,172.30.0.1,qe-zhsun-999-manualmaster-etcd-1,qe-zhsun-999-manualnrr-1" "http_proxy" "http://file.rdu.redhat.com:3128" "https_proxy" "http://file.rdu.redhat.com:3128" "no_proxy" ".centralci.eng.rdu2.redhat.com,.cluster.local,.svc,169.254.169.254,172.16.120.62,172.30.0.1,qe-zhsun-999-manualmaster-etcd-1,qe-zhsun-999-manualnrr-1" ---> Running in 150bb9c3aa88 ---> f276df8b890f Removing intermediate container 150bb9c3aa88 Step 3/9 : LABEL "com.redhat.apb.spec" "LS0tCnZlcnNpb246IDEuMApuYW1lOiBoZWxsby13b3JsZC1hcGIKZGVzY3JpcHRpb246IGRlcGxveXMgaGVsbG8td29ybGQgd2ViIGFwcGxpY2F0aW9uCmJpbmRhYmxlOiAiRmFsc2UiCmFzeW5jOiBvcHRpb25hbAptZXRhZGF0YToKICBkaXNwbGF5TmFtZTogSGVsbG8gV29ybGQgKEFQQikKICBsb25nRGVzY3JpcHRpb246CiAgICBBIHNhbXBsZSBBUEIgd2hpY2ggZGVwbG95cyBhIGNvbnRhaW5lcml6ZWQgSGVsbG8gV29ybGQgd2ViIGFwcGxpY2F0aW9uCiAgZGVwZW5kZW5jaWVzOiBbJ2RvY2tlci5pby9hbnNpYmxlcGxheWJvb2tidW5kbGUvaGVsbG8td29ybGQ6bGF0ZXN0J10KICBwcm92aWRlckRpc3BsYXlOYW1lOiAiUmVkIEhhdCwgSW5jLiIKcGxhbnM6CiAgLSBuYW1lOiBkZWZhdWx0CiAgICBkZXNjcmlwdGlvbjogQSBzYW1wbGUgQVBCIHdoaWNoIGRlcGxveXMgSGVsbG8gV29ybGQKICAgIGZyZWU6ICJUcnVlIgogICAgbWV0YWRhdGE6CiAgICAgIGRpc3BsYXlOYW1lOiBEZWZhdWx0CiAgICAgIGxvbmdEZXNjcmlwdGlvbjoKICAgICAgICBUaGlzIHBsYW4gZGVwbG95cyBhIFB5dGhvbiB3ZWIgYXBwbGljYXRpb24gZGlzcGxheWluZyBIZWxsbyBXb3JsZAogICAgICBjb3N0OiAkMC4wMAogICAgcGFyYW1ldGVyczogW10K" ---> Running in 1554a25d0970 ---> 13a7e4c5b342 Removing intermediate container 1554a25d0970 Step 4/9 : ADD playbooks /opt/apb/actions ---> 491c7cc16f05 Removing intermediate container 8106488f7e69 Step 5/9 : ADD . /opt/ansible/roles/hello-world-apb ---> cdebbf4ebd01 Removing intermediate container cf6754f25c71 Step 6/9 : RUN chmod -R g=u /opt/{ansible,apb} ---> Running in d0acd25f12f3 ---> 1b316a658bc8 Removing intermediate container d0acd25f12f3 Step 7/9 : USER apb ---> Running in 0de84630a189 ---> 687bd041e538 Removing intermediate container 0de84630a189 Step 8/9 : ENV "OPENSHIFT_BUILD_NAME" "hellooooo-1" "OPENSHIFT_BUILD_NAMESPACE" "openshift" "OPENSHIFT_BUILD_SOURCE" "https://github.com/ansibleplaybookbundle/hello-world-apb.git" "OPENSHIFT_BUILD_REFERENCE" "master" "OPENSHIFT_BUILD_COMMIT" "7132053ccc8ea64b856074247638cfde54feaaff" ---> Running in d8642ea92a71 ---> 3f1d962b12ac Removing intermediate container d8642ea92a71 Step 9/9 : LABEL "io.openshift.build.commit.author" "David Zager \u003cdzager\u003e" "io.openshift.build.commit.date" "Thu Apr 12 12:06:42 2018 -0400" "io.openshift.build.commit.id" "7132053ccc8ea64b856074247638cfde54feaaff" "io.openshift.build.commit.message" "Skip last operation if not in cluster (#6)" "io.openshift.build.commit.ref" "master" "io.openshift.build.name" "hellooooo-1" "io.openshift.build.namespace" "openshift" "io.openshift.build.source-location" "https://github.com/ansibleplaybookbundle/hello-world-apb.git" ---> Running in 7105edef86a3 ---> 7847b40b76ee Removing intermediate container 7105edef86a3 Successfully built 7847b40b76ee Pushing image docker-registry.default.svc:5000/openshift/hellooooo:latest ... Pushed 0/17 layers, 0% complete Pushed 1/17 layers, 29% complete Pushed 2/17 layers, 29% complete Pushed 3/17 layers, 29% complete Pushed 4/17 layers, 47% complete Pushed 5/17 layers, 47% complete Pushed 6/17 layers, 59% complete Pushed 7/17 layers, 59% complete Pushed 8/17 layers, 59% complete Pushed 9/17 layers, 54% complete Pushed 10/17 layers, 60% complete Pushed 11/17 layers, 67% complete Pushed 12/17 layers, 73% complete Pushed 13/17 layers, 78% complete Pushed 14/17 layers, 86% complete Pushed 15/17 layers, 92% complete Pushed 16/17 layers, 96% complete Pushed 16/17 layers, 100% complete Pushed 17/17 layers, 100% complete Push successful [szh@localhost hello-world-apb]$ oc get images | grep hellooooo sha256:ed059ac605dde64fd4be9e3af8b6a761f4d9cf5f240ab38e2245f0a7c575b0e6 docker-registry.default.svc:5000/openshift/hellooooo@sha256:ed059ac605dde64fd4be9e3af8b6a761f4d9cf5f240ab38e2245f0a7c575b0e6 2) use my own apb couldn't get image. [szh@localhost tes]$ apb init my-999-apb Initializing /home/szh/code/tes/my-999-apb for an APB. Generating playbook files Successfully initialized project directory at: /home/szh/code/tes/my-999-apb Please run *apb prepare* inside of this directory after editing files. [szh@localhost tes]$ cd my-999-apb/ [szh@localhost my-999-apb]$ apb prepare Finished writing dockerfile. [szh@localhost my-999-apb]$ oc new-app ./ --name my-999-apb -n openshift --> Found Docker image 57595a3 (3 weeks old) from Docker Hub for "ansibleplaybookbundle/apb-base" * An image stream will be created as "apb-base:latest" that will track the source image * A Docker build using binary input will be created * The resulting image will be pushed to image stream "my-999-apb:latest" * A binary build was created, use 'start-build --from-dir' to trigger a new build * This image will be deployed in deployment config "my-999-apb" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/my-999-apb --port=[port]' later * WARNING: Image "ansibleplaybookbundle/apb-base" runs as the 'root' user which may not be permitted by your cluster administrator --> Creating resources ... imagestream "my-999-apb" created buildconfig "my-999-apb" created deploymentconfig "my-999-apb" created --> Success Build scheduled, use 'oc logs -f bc/my-999-apb' to track its progress. Run 'oc status' to view your app. [szh@localhost my-999-apb]$ oc logs -f bc/my-999-apb -n openshift error: no builds found for "my-999-apb" [szh@localhost my-999-apb]$ oc get images | grep apb-base sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b ansibleplaybookbundle/apb-base@sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b [szh@localhost my-999-apb]$ oc get is --all-namespaces | grep my-999-apb openshift my-999-apb docker-registry.default.svc:5000/openshift/my-999-apb [szh@localhost my-999-apb]$ oc get bc --all-namespaces | grep my-999-apb openshift my-999-apb Docker Binary 0 [szh@localhost my-999-apb]$ oc get images | grep my-999-apb [szh@localhost my-999-apb]$
"oc run" failed, couldn't pull image. [szh@localhost hello-world-apb]$ oc new-project apb-project Now using project "apb-project" on server "https://host-8-243-101.host.centralci.eng.rdu2.redhat.com:8443". [szh@localhost hello-world-apb]$ oc create serviceaccount apb serviceaccount "apb" created [szh@localhost hello-world-apb]$ oc create rolebinding apb --clusterrole=admin --serviceaccount=apb-project:apb rolebinding "apb" created [szh@localhost hello-world-apb]$ oc run pod-hello \ > --env="POD_NAME=pod-hello" \ > --env="POD_NAMESPACE=apb-project" \ > --image=172.30.46.231:5000/openshift/hellooooo \ > --restart=Never \ > --attach=true \ > --serviceaccount=apb \ > -- provision -e namespace=apb-project -e cluster=$CLUSTER [szh@localhost hello-world-apb]$ oc get pods -n apb-project NAME READY STATUS RESTARTS AGE pod-hello 0/1 ImagePullBackOff 0 59m [szh@localhost hello-world-apb]$oc describe pod pod-hello Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 1h default-scheduler Successfully assigned pod-hello to qe-zhsun-999-manualnrr-1 Normal SandboxChanged 1h (x13 over 1h) kubelet, qe-zhsun-999-manualnrr-1 Pod sandbox changed, it will be killed and re-created. Normal Pulling 1h (x3 over 1h) kubelet, qe-zhsun-999-manualnrr-1 pulling image "172.30.46.231:5000/openshift/hellooooo" Warning Failed 1h (x3 over 1h) kubelet, qe-zhsun-999-manualnrr-1 Failed to pull image "172.30.46.231:5000/openshift/hellooooo": rpc error: code = Unknown desc = Get https://172.30.46.231:5000/v1/_ping: Forbidden Warning Failed 1h (x3 over 1h) kubelet, qe-zhsun-999-manualnrr-1 Error: ErrImagePull Normal BackOff 6m (x250 over 1h) kubelet, qe-zhsun-999-manualnrr-1 Back-off pulling image "172.30.46.231:5000/openshift/hellooooo" Warning Failed 1m (x270 over 1h) kubelet, qe-zhsun-999-manualnrr-1 Error: ImagePullBackOff [root@qe-zhsun-999-manualmaster-etcd-1 ~]# docker pull 172.30.46.231:5000/openshift/hellooooo Using default tag: latest Trying to pull repository 172.30.46.231:5000/openshift/hellooooo ... Pulling repository 172.30.46.231:5000/openshift/hellooooo
From what I can tell the problem is that your image is named: docker-registry.default.svc:5000/openshift/hellooooo Yet in `oc run` you used: --image 172.30.46.231:5000/openshift/hellooooo I would specify the name of the image you find when you do `oc get images | grep hellooo` since the registry will determine the FQName. Alex, I would include this note in the documentation and I can submit a note upstream as well.
Updated per comment 48: https://github.com/openshift/openshift-docs/pull/9648 @Dylan, PTAL
Verified failed. oc run failed. [szh@localhost hello-world-apb]$ oc get images |grep hello sha256:37ba76423a825a5902ddbe36baf0f2615ddb15e6220ebd994a28727cc0424c16 docker-registry.default.svc:5000/openshift/hello@sha256:37ba76423a825a5902ddbe36baf0f2615ddb15e6220ebd994a28727cc0424c16 [szh@localhost hello-world-apb]$ oc run pod-hello --env="POD_NAME=pod-hello" --env="POD_NAMESPACE=apb-project" --image=docker-registry.default.svc:5000/openshift/hello --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=apb-project -e cluster=$CLUSTER If you don't see a command prompt, try pressing enter. [DEPRECATION WARNING]: openshift_raw is kept for backwards compatibility but usage is discouraged. The module documentation details page may explain more about this rationale.. This feature will be removed in a future release. Deprecation warnings can be disabled by setting deprecation_warnings=False in ansible.cfg. [DEPRECATION WARNING]: k8s_raw is kept for backwards compatibility but usage is discouraged. The module documentation details page may explain more about this rationale.. This feature will be removed in a future release. Deprecation warnings can be disabled by setting deprecation_warnings=False in ansible.cfg. [WARNING]: Found variable using reserved name: name PLAY [hello-world-apb provision] *********************************************** TASK [ansibleplaybookbundle.asb-modules : debug] ******************************* skipping: [localhost] TASK [hello-world-apb : Update last operation] ********************************* fatal: [localhost]: FAILED! => {"changed": false, "msg": "Error attempting to update pod with last operation annotation: (403)\nReason: Forbidden\nHTTP response headers: HTTPHeaderDict({'Date': 'Fri, 01 Jun 2018 05:31:59 GMT', 'Content-Length': '303', 'Content-Type': 'application/json', 'Cache-Control': 'no-store'})\nHTTP response body: {\n \"kind\": \"Status\",\n \"apiVersion\": \"v1\",\n \"metadata\": {},\n \"status\": \"Failure\",\n \"message\": \"pods \\\"pod-hello\\\" is forbidden: unable to validate against any security context constraint: []\",\n \"reason\": \"Forbidden\",\n \"details\": {\n \"name\": \"pod-hello\",\n \"kind\": \"pods\"\n },\n \"code\": 403\n}\n"} PLAY RECAP ********************************************************************* localhost : ok=0 changed=0 unreachable=0 failed=1 pod apb-project/pod-hello terminated (Error) [szh@localhost hello-world-apb]$ oc get pod -n apb-project NAME READY STATUS RESTARTS AGE pod-hello 0/1 Error 0 32m [szh@localhost hello-world-apb]$ oc describe pod -n apb-project ... Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 10m default-scheduler Successfully assigned pod-hello to qe-zhsun-gceenode-registry-router-1 Normal Pulled 10m kubelet, qe-zhsun-gceenode-registry-router-1 Container image "docker-registry.default.svc:5000/openshift/hello" already present on machine Normal Created 10m kubelet, qe-zhsun-gceenode-registry-router-1 Created container Normal Started 10m kubelet, qe-zhsun-gceenode-registry-router-1 Started container
This means the actual `oc run` succeeded but the hello-world-apb expects you to pass in an extra variable when running outside the context of the broker. You have two options: 1. run the exact same `oc run` command with an extra variable passed in: oc run pod-hello --env="POD_NAME=pod-hello" --env="POD_NAMESPACE=apb-project" --image=docker-registry.default.svc:5000/openshift/hello --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=apb-project -e cluster=$CLUSTER -e in_cluster=false 2. You can simply not pass in POD_NAME and POD_NAMESPACE as env vars. I added those environment variables to the documentation because that is what the broker is doing so if an APB needs those variables it is supplied. However in the hello-world-apb case its setting in_cluster=true when POD_NAME and POD_NAMESPACE is defined. I can submit some changes to get the last_operation command disabled by default but this bug should be verified. This is a problem with the variables being supplied to the APB and not with executing the pod in the cluster. Any APB that does not use the `asb_last_operation` task will succeed with these instructions. The `asb_last_operation` task is broker dependent so we should not be executing that task by default.
Commit pushed to master at https://github.com/openshift/openshift-docs https://github.com/openshift/openshift-docs/commit/fc8bc12f5dcbc94557e5cf2ae08e0f87094336d1 Merge pull request #9648 from adellape/apb_push Bug 1526147: Add 'Working with Remote Clusters' APB doc
Dylan, Sorry for that I couldn't verify this bug. 1) I use hello-world-apb as example. $ oc run pod-hello7 --env="POD_NAME=pod-hello7" --env="POD_NAMESPACE=apb-project" --image=docker-registry.default.svc:5000/openshift/hello --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=apb-project -e cluster=$CLUSTER -e in_cluster=false fatal: [localhost]: FAILED! => {"ansible_facts": {"deployment": [], "deployment_config": [], "route": [], "service": {"apiVersion": "v1", "kind": "Service", "metadata": {"creationTimestamp": "2018-06-04T07:57:29Z", "labels": {"apb_name": "hello-world", "apb_plan_id": "default", "apb_service_class_id": "0", "apb_service_instance_id": "0"}, "name": "hello-world-0", "namespace": "apb-project", "resourceVersion": "5526", "selfLink": "/api/v1/namespaces/apb-project/services/hello-world-0", "uid": "ed7a94cb-67cc-11e8-bdbe-fa163ea59861"}, "spec": {"clusterIP": "172.30.0.153", "ports": [{"name": "web", "port": 8080, "protocol": "TCP", "targetPort": 8080}], "selector": {"app": "hello-world-0", "service": "hello-world-0"}, "sessionAffinity": "None", "type": "ClusterIP"}, "status": {"loadBalancer": {}}}}, "attempts": 10, "changed": false} PLAY RECAP ********************************************************************* localhost : ok=2 changed=0 unreachable=0 failed=1 pod apb-project/pod-hello7 terminated (Error) $ oc run pod-hello8 --image=docker-registry.default.svc:5000/openshift/hello --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=apb-project -e cluster=$CLUSTER -e in_cluster=false If you don't see a command prompt, try pressing enter. fatal: [localhost]: FAILED! => {"ansible_facts": {"deployment": [], "deployment_config": [], "route": [], "service": {"apiVersion": "v1", "kind": "Service", "metadata": {"creationTimestamp": "2018-06-04T07:57:29Z", "labels": {"apb_name": "hello-world", "apb_plan_id": "default", "apb_service_class_id": "0", "apb_service_instance_id": "0"}, "name": "hello-world-0", "namespace": "apb-project", "resourceVersion": "5526", "selfLink": "/api/v1/namespaces/apb-project/services/hello-world-0", "uid": "ed7a94cb-67cc-11e8-bdbe-fa163ea59861"}, "spec": {"clusterIP": "172.30.0.153", "ports": [{"name": "web", "port": 8080, "protocol": "TCP", "targetPort": 8080}], "selector": {"app": "hello-world-0", "service": "hello-world-0"}, "sessionAffinity": "None", "type": "ClusterIP"}, "status": {"loadBalancer": {}}}}, "attempts": 10, "changed": false} PLAY RECAP ********************************************************************* localhost : ok=2 changed=0 unreachable=0 failed=1 pod apb-project/pod-hello8 terminated (Error) $ oc get pod -n apb-project NAME READY STATUS RESTARTS AGE pod-hello7 0/1 Error 0 11m pod-hello8 0/1 Error 0 3m 2）I use my own apb as example. After run "apb init" and "apb prepare" .run "oc new-app ./ --name my-07 -n openshift" success. But "oc get images | grep my-07" couldn't get image. $ oc logs -f bc/my-07 -n openshift error: no builds found for "my-07" $ oc get images | grep my-07 Base on this doc I couldn't get completely nomal result. Could you provide an apb example in the doc? Thanks.
Sunzhaohua, I just tested this on my cluster using the hello-world-apb from our dockerhub org with no problem. This is the command I run: $ oc run hello8 --image=docker.io/ansibleplaybookbundle/hello-world-apb --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=dylan-test -e cluster=openshift -e in_cluster=false I'm not sure what the issue you are hitting is but the main thing I noticed is that you are setting `cluster=$CLUSTER`... what is $CLUSTER set to? It should be `openshift`. I also am not sure by your comment what task actually failed. The above command ran all of the APB actions without error. Please make sure you are passing in the same values that I am above and ensure you are using the latest hello-world example.
verified. I think we'd better update "cluster=$CLUSTER" to "cluster=openshift" [szh@localhost hello-world-apb]$ oc new-app ./ --name hello -n openshift --> Found Docker image 486a293 (13 days old) from Docker Hub for "ansibleplaybookbundle/apb-base:canary" * An image stream will be created as "apb-base:canary" that will track the source image * A Docker build using source code from https://github.com/ansibleplaybookbundle/hello-world-apb.git#master will be created * The resulting image will be pushed to image stream "hello:latest" * Every time "apb-base:canary" changes a new build will be triggered * This image will be deployed in deployment config "hello" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/hello --port=[port]' later * WARNING: Image "ansibleplaybookbundle/apb-base:canary" runs as the 'root' user which may not be permitted by your cluster administrator --> Creating resources ... imagestream "apb-base" created imagestream "hello" created buildconfig "hello" created deploymentconfig "hello" created --> Success Build scheduled, use 'oc logs -f bc/hello' to track its progress. Run 'oc status' to view your app. [szh@localhost hello-world-apb]$ oc logs -f bc/hello -n openshift Cloning "https://github.com/ansibleplaybookbundle/hello-world-apb.git" ... Commit: aacda7eb2d90307f04090fb1b1f68bf8769fa5ab (Update hello-world-apb for ansible 2.6 (#8)) Author: David Zager <dzager> Date: Mon Jun 4 13:17:42 2018 -0400 Pulling image ansibleplaybookbundle/apb-base@sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b ... Pulled 0/14 layers, 0% complete Pulled 1/14 layers, 7% complete Pulled 2/14 layers, 23% complete Pulled 3/14 layers, 31% complete Pulled 4/14 layers, 39% complete Pulled 5/14 layers, 49% complete Pulled 6/14 layers, 56% complete Pulled 7/14 layers, 64% complete Pulled 8/14 layers, 64% complete Pulled 9/14 layers, 71% complete Pulled 10/14 layers, 71% complete Pulled 11/14 layers, 79% complete Pulled 12/14 layers, 86% complete Pulled 13/14 layers, 93% complete Pulled 14/14 layers, 100% complete Extracting Step 1/8 : FROM ansibleplaybookbundle/apb-base@sha256:a501e750921fad25f8ef3a498a72694722ad5b11457d989eaefacf854f44d41b ---> 486a2936c6a8 Step 2/8 : LABEL "com.redhat.apb.spec" "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" ---> Running in a0c8ac6bd74d ---> d1dea285e6e5 Removing intermediate container a0c8ac6bd74d Step 3/8 : ADD playbooks /opt/apb/actions ---> 008b06b449b0 Removing intermediate container d3a0b08089ed Step 4/8 : ADD . /opt/ansible/roles/hello-world-apb ---> 68023e8fe870 Removing intermediate container 21844d3f268a Step 5/8 : RUN chmod -R g=u /opt/{ansible,apb} ---> Running in ed926dfb7084 ---> e4bd36d27a97 Removing intermediate container ed926dfb7084 Step 6/8 : USER apb ---> Running in 3fb6212120d3 ---> 3b4ed71f0d98 Removing intermediate container 3fb6212120d3 Step 7/8 : ENV "OPENSHIFT_BUILD_NAME" "hello-1" "OPENSHIFT_BUILD_NAMESPACE" "openshift" "OPENSHIFT_BUILD_SOURCE" "https://github.com/ansibleplaybookbundle/hello-world-apb.git" "OPENSHIFT_BUILD_REFERENCE" "master" "OPENSHIFT_BUILD_COMMIT" "aacda7eb2d90307f04090fb1b1f68bf8769fa5ab" ---> Running in 97075ffaa590 ---> b1215c2c79b1 Removing intermediate container 97075ffaa590 Step 8/8 : LABEL "io.openshift.build.commit.author" "David Zager \u003cdzager\u003e" "io.openshift.build.commit.date" "Mon Jun 4 13:17:42 2018 -0400" "io.openshift.build.commit.id" "aacda7eb2d90307f04090fb1b1f68bf8769fa5ab" "io.openshift.build.commit.message" "Update hello-world-apb for ansible 2.6 (#8)" "io.openshift.build.commit.ref" "master" "io.openshift.build.name" "hello-1" "io.openshift.build.namespace" "openshift" "io.openshift.build.source-location" "https://github.com/ansibleplaybookbundle/hello-world-apb.git" ---> Running in b54da26aeac5 ---> 1a2d388357c2 Removing intermediate container b54da26aeac5 Successfully built 1a2d388357c2 Pushing image docker-registry.default.svc:5000/openshift/hello:latest ... Pushed 0/17 layers, 0% complete Pushed 1/17 layers, 6% complete Pushed 2/17 layers, 12% complete Pushed 3/17 layers, 18% complete Pushed 4/17 layers, 24% complete Pushed 5/17 layers, 32% complete Pushed 6/17 layers, 38% complete Pushed 7/17 layers, 54% complete Pushed 8/17 layers, 60% complete Pushed 9/17 layers, 71% complete Pushed 10/17 layers, 76% complete Pushed 11/17 layers, 82% complete Pushed 12/17 layers, 88% complete Pushed 13/17 layers, 94% complete Pushed 14/17 layers, 100% complete Pushed 15/17 layers, 100% complete Pushed 16/17 layers, 100% complete Pushed 17/17 layers, 100% complete Push successful [szh@localhost hello-world-apb]$ oc get images | grep hello sha256:8e01328b77844c02977d82342c9b66cf6f57258b46446263d3efce3aec48b785 docker-registry.default.svc:5000/openshift/hello@sha256:8e01328b77844c02977d82342c9b66cf6f57258b46446263d3efce3aec48b785 [szh@localhost hello-world-apb]$ oc new-project apb-project Now using project "apb-project" on server "https://host-8-250-138.host.centralci.eng.rdu2.redhat.com:8443". [szh@localhost hello-world-apb]$ oc create rolebinding apb --clusterrole=admin --serviceaccount=apb-project:apb rolebinding "apb" created [szh@localhost hello-world-apb]$ oc create serviceaccount apb serviceaccount "apb" created [szh@localhost hello-world-apb]$ oc run hello --image=docker.io/ansibleplaybookbundle/hello-world-apb --restart=Never --attach=true --serviceaccount=apb -- provision -e namespace=apb-project -e cluster=openshift -e in_cluster=false If you don't see a command prompt, try pressing enter. changed: [localhost] TASK [hello-world-apb : route state=present] *********************************** changed: [localhost] TASK [hello-world-apb : Update last operation] ********************************* skipping: [localhost] TASK [hello-world-apb : include_tasks] ***************************************** included: /opt/ansible/roles/hello-world-apb/tasks/verify_provision.yml for localhost TASK [hello-world-apb : Verify hello-world-0 objects exist] ******************** ok: [localhost] TASK [hello-world-apb : Wait for deployment to become available] *************** skipping: [localhost] TASK [hello-world-apb : Wait for deployment config to become available] ******** FAILED - RETRYING: Wait for deployment config to become available (12 retries left).Result was: { "attempts": 1, "changed": false, "msg": "DeploymentConfig available status: False", "retries": 13 } ok: [localhost] => { "attempts": 2, "msg": "DeploymentConfig available status: False" } TASK [hello-world-apb : Update last operation] ********************************* skipping: [localhost] PLAY RECAP ********************************************************************* localhost : ok=6 changed=3 unreachable=0 failed=0 [szh@localhost hello-world-apb]$ oc get pod NAME READY STATUS RESTARTS AGE hello 0/1 Completed 0 3m hello-world-0-1-sptnd 1/1 Running 0 2m
Thanks for retesting! I submitted a PR so that there is no confusion over the CLUSTER param: https://github.com/openshift/openshift-docs/pull/9850
Commits pushed to master at https://github.com/openshift/openshift-docs https://github.com/openshift/openshift-docs/commit/378478eed03616849997754f204e893c305c359d Bug 1526147 - Update oc run instructions to include cluster name https://github.com/openshift/openshift-docs/commit/23d4a9eee7c16a750cfd7be05e388bd30d89df44 Merge pull request #9850 from dymurray/open Bug 1526147 - Update oc run instructions to include cluster name
Also having error when trying todo `apb test` Steps: 1. git clone git:aerogearcatalog/aerogear-digger-apb.git 2. oc new-app aerogear-digger-apb --name aerogear-digger-apb -n openshift 3. oc logs -f bc/aerogear-digger-apb -n openshift info: Logs available at https://jenkins-openshift.192.168.37.1.nip.io/blue/organizations/jenkins/openshift%2Fopenshift-aerogear-digger-apb/detail/openshift-aerogear-digger-apb/1/ 4. open https://jenkins-openshift.192.168.37.1.nip.io/blue/organizations/jenkins/openshift%2Fopenshift-aerogear-digger-apb/detail/openshift-aerogear-digger-apb/1/ Output: OpenShift Build openshift/aerogear-digger-apb-1 from git:aerogearcatalog/aerogear-digger-apb.git Checking out git git:aerogearcatalog/aerogear-digger-apb.git into /var/lib/jenkins/jobs/openshift/jobs/openshift-aerogear-digger-apb/workspace@script to read Jenkinsfile Cloning the remote Git repository Cloning repository git:aerogearcatalog/aerogear-digger-apb.git > git init /var/lib/jenkins/jobs/openshift/jobs/openshift-aerogear-digger-apb/workspace@script # timeout=10 Fetching upstream changes from git:aerogearcatalog/aerogear-digger-apb.git > git --version # timeout=10 > git fetch --tags --progress git:aerogearcatalog/aerogear-digger-apb.git +refs/heads/*:refs/remotes/origin/* ERROR: Error cloning remote repo 'origin' hudson.plugins.git.GitException: Command "git fetch --tags --progress git:aerogearcatalog/aerogear-digger-apb.git +refs/heads/*:refs/remotes/origin/*" returned status code 128: stdout: stderr: Host key verification failed. fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists. at org.jenkinsci.plugins.gitclient.CliGitAPIImpl.launchCommandIn(CliGitAPIImpl.java:1990) at org.jenkinsci.plugins.gitclient.CliGitAPIImpl.launchCommandWithCredentials(CliGitAPIImpl.java:1709) at org.jenkinsci.plugins.gitclient.CliGitAPIImpl.access$300(CliGitAPIImpl.java:72) at org.jenkinsci.plugins.gitclient.CliGitAPIImpl$1.execute(CliGitAPIImpl.java:400) at org.jenkinsci.plugins.gitclient.CliGitAPIImpl$2.execute(CliGitAPIImpl.java:609) at hudson.plugins.git.GitSCM.retrieveChanges(GitSCM.java:1120) at hudson.plugins.git.GitSCM.checkout(GitSCM.java:1160) at org.jenkinsci.plugins.workflow.steps.scm.SCMStep.checkout(SCMStep.java:109) at org.jenkinsci.plugins.workflow.cps.CpsScmFlowDefinition.create(CpsScmFlowDefinition.java:130) at org.jenkinsci.plugins.workflow.cps.CpsScmFlowDefinition.create(CpsScmFlowDefinition.java:59) at org.jenkinsci.plugins.workflow.job.WorkflowRun.run(WorkflowRun.java:263) at hudson.model.ResourceController.execute(ResourceController.java:97) at hudson.model.Executor.run(Executor.java:429) ERROR: Error cloning remote repo 'origin' Finished: FAILURE
Sunzhaohua, are you experiencing the issue cited in comment 58 using the latest doc here? https://docs.openshift.org/latest/apb_devel/writing/reference.html#apb-devel-writing-ref-remote-clusters Dylan, any ideas?
Alex, I test hello-world-apb and it works well, refer to commnet 55. I test your apb also get errors but not same with you. 1. $ git clone https://github.com/aerogearcatalog/aerogear-digger-apb.git 2. $ cat apb.yml | base64 copy and paste result into the Dockerfile 3. $ oc new-app aerogear-digger-apb --name aerogear-digger-apb -n openshift * A pipeline build using source code from https://github.com/aerogearcatalog/aerogear-digger-apb.git#master will be created * Use 'start-build' to trigger a new build --> Creating resources ... buildconfig "aerogear-digger-apb" created --> Success Build scheduled, use 'oc logs -f bc/aerogear-digger-apb' to track its progress. Run 'oc status' to view your app. 4. $ oc logs -f bc/aerogear-digger-apb -n openshift info: Logs available at https://jenkins-openshift.apps.0628-9fi.qe.rhcloud.com/blue/organizations/jenkins/openshift%2Fopenshift-aerogear-digger-apb/detail/openshift-aerogear-digger-apb/1/ open https://jenkins-openshift.apps.0628-9fi.qe.rhcloud.com/blue/organizations/jenkins/openshift%2Fopenshift-aerogear-digger-apb/detail/openshift-aerogear-digger-apb/1/ Output: OpenShift Build openshift/aerogear-digger-apb-1 from https://github.com/aerogearcatalog/aerogear-digger-apb.git Logs OpenShift Build openshift/aerogear-digger-apb-1 from https://github.com/aerogearcatalog/aerogear-digger-apb.git Checking out git https://github.com/aerogearcatalog/aerogear-digger-apb.git into /var/lib/jenkins/jobs/openshift/jobs/openshift-aerogear-digger-apb/workspace@script to read Jenkinsfile Cloning the remote Git repository Cloning repository https://github.com/aerogearcatalog/aerogear-digger-apb.git > git init /var/lib/jenkins/jobs/openshift/jobs/openshift-aerogear-digger-apb/workspace@script # timeout=10 Fetching upstream changes from https://github.com/aerogearcatalog/aerogear-digger-apb.git > git --version # timeout=10 > git fetch --tags --progress https://github.com/aerogearcatalog/aerogear-digger-apb.git +refs/heads/*:refs/remotes/origin/* > git config remote.origin.url https://github.com/aerogearcatalog/aerogear-digger-apb.git # timeout=10 > git config --add remote.origin.fetch +refs/heads/*:refs/remotes/origin/* # timeout=10 > git config remote.origin.url https://github.com/aerogearcatalog/aerogear-digger-apb.git # timeout=10 Fetching upstream changes from https://github.com/aerogearcatalog/aerogear-digger-apb.git > git fetch --tags --progress https://github.com/aerogearcatalog/aerogear-digger-apb.git +refs/heads/*:refs/remotes/origin/* > git rev-parse origin/master^{commit} # timeout=10 Checking out Revision baf7479751448673e1a8cdb3acd0a0446137a3df (origin/master) > git config core.sparsecheckout # timeout=10 > git checkout -f baf7479751448673e1a8cdb3acd0a0446137a3df Commit message: "Merge pull request #27 from philbrookes/add-labels" First time build. Skipping changelog. ERROR: Could not find any definition of libraries [fh-pipeline-library] org.codehaus.groovy.control.MultipleCompilationErrorsException: startup failed: WorkflowScript: Loading libraries failed 1 error at org.codehaus.groovy.control.ErrorCollector.failIfErrors(ErrorCollector.java:310) at org.codehaus.groovy.control.CompilationUnit.applyToPrimaryClassNodes(CompilationUnit.java:1085) at org.codehaus.groovy.control.CompilationUnit.doPhaseOperation(CompilationUnit.java:603) at org.codehaus.groovy.control.CompilationUnit.processPhaseOperations(CompilationUnit.java:581) at org.codehaus.groovy.control.CompilationUnit.compile(CompilationUnit.java:558) at groovy.lang.GroovyClassLoader.doParseClass(GroovyClassLoader.java:298) at groovy.lang.GroovyClassLoader.parseClass(GroovyClassLoader.java:268) at groovy.lang.GroovyShell.parseClass(GroovyShell.java:688) at groovy.lang.GroovyShell.parse(GroovyShell.java:700) at org.jenkinsci.plugins.workflow.cps.CpsGroovyShell.doParse(CpsGroovyShell.java:129) at org.jenkinsci.plugins.workflow.cps.CpsGroovyShell.reparse(CpsGroovyShell.java:123) at org.jenkinsci.plugins.workflow.cps.CpsFlowExecution.parseScript(CpsFlowExecution.java:517) at org.jenkinsci.plugins.workflow.cps.CpsFlowExecution.start(CpsFlowExecution.java:480) at org.jenkinsci.plugins.workflow.job.WorkflowRun.run(WorkflowRun.java:269) at hudson.model.ResourceController.execute(ResourceController.java:97) at hudson.model.Executor.run(Executor.java:429) Finished: FAILURE 1.4.1 · Core 2.107.3 · 61988a2 · 9th February 2018 02:26 AM
Dylan, I re-test some apbs. For the pushing APBs, 1) I using rhpam-apb(https://github.com/ansibleplaybookbundle/rhpam-apb.git), import-vm-apb(https://github.com/ansibleplaybookbundle/import-vm-apb.git). It works well. 2) If I using mysql-apb(https://github.com/ansibleplaybookbundle/mysql-apb.git) it return "no language matched" $ oc new-app ./ --name mysql-apb -n openshift error: No language matched the source repository 3) If I first run `apb init my-01-apb`, `apb prepare`,then run `oc new-app ./ --name my-01-apb -n openshift`, I couldn't get image. $ oc logs -f bc/my-01-apb -n openshift error: no builds found for "my-01-apb" I don't know the difference between these apbs, could you help to confirm? If I run `apb push` in cluster, these apbs will be pushed successful. I change the status to assigned, if I am wrong, please correct me. Thanks.
sunzhaohua, mysql-apb repo has 3 dockerfiles `Dockerfile-canary, Dockerfile-latest, Dockerfile-nightly`. This is to help us in the build process. `oc new-app` expects the dockerfile to simply be named `Dockerfile` like it is in rhpam-apb & import-vm-apb. This is because source-to-image uses a `docker` strategy which it can only know to use if it finds a `Dockerfile` in the repo. For 3. I do not have enough information as to why it failed. I would refer you to the documentation for `oc new-app` to get a better understanding of the s2i functionality. Another option is to simply do `oc new-build . --to <name>` to test the build step out manually. https://docs.openshift.com/enterprise/3.1/dev_guide/new_app.html
@sunzhaohua, ahh build fails because we use pipeline library in Jenkinsfile, that needs additional configuration for Jenkins. Thanks for testing it
Does this still need to be on ASSIGNED?
Alex, As Dylan's explanation, I think it's ok to me.
As Dylan's explanation, and the updated doc according to this pr: https://github.com/openshift/openshift-docs/pull/9648. Verified this bug.
https://docs.openshift.com/container-platform/3.10/apb_devel/writing/reference.html#apb-devel-writing-ref-remote-clusters
*** Bug 1557462 has been marked as a duplicate of this bug. ***