openshift-ansible: Failed to install Openshift 3.10 because of missing packages
Description
Failed to install Openshift 3.10 because of missing packages.
Version
ansible 2.5.2
openshift-ansible-3.10.0-0.41.0
OS - Centos 7.5
Steps To Reproduce
Run deploy_cluster.yml playbook
Expected Results
Openshift origin 3.10 installed successfully
Observed Results
TASK [openshift_node : Install node, clients, and conntrack packages] **********
Wednesday 16 May 2018 15:57:47 +0000 (0:00:02.764) 0:10:00.906 *********
FAILED - RETRYING: Install node, clients, and conntrack packages (3 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (3 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (3 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (2 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (2 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (2 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (1 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (1 retries left).
FAILED - RETRYING: Install node, clients, and conntrack packages (1 retries left).
failed: [lago-node0] (item={u'name': u'origin-node-3.10*'}) => {
"attempts": 3,
"changed": false,
"item": {
"name": "origin-node-3.10*"
},
"rc": 126,
"results": [
"No package matching 'origin-node-3.10*' found available, installed or updated"
]
}
No package matching 'origin-clients-3.10*' found available, installed or updated
failed: [lago-node0] (item={u'name': u'origin-clients-3.10*'}) => {
"attempts": 3,
"changed": false,
"item": {
"name": "origin-clients-3.10*"
},
"rc": 126,
"results": [
"No package matching 'origin-clients-3.10*' found available, installed or updated"
]
}
Additional Information
-
I also didn’t see that origin--3.10 is available in the testing repo: https://buildlogs.centos.org/centos/7/paas/x86_64/openshift-origin/
-
I don’t understand why installing any RPM packages on the masters/nodes when
containerized=true
About this issue
- Original URL
- State: closed
- Created 6 years ago
- Comments: 90 (43 by maintainers)
in case you haven’t seen the notification sent to dev mailing list
Let me know if there are any issues either here or writing to the mailer
@alv91 correct because we had to remove the package yesterday evening. And so the expectation is that everyone who already have a non pristine deployment or an env where the
python2-docker-3.3.0-1.el7.noarch.rpm&python-docker-py >= 1.7.2-1are present to uninstall itAfter checking out
openshift-ansible-3.10.27-1tag thisatomicissue didn’t happen.guys i understand the excitement but it doesn’t help if everyone is adding a comment on top of another comment.
the reason i haven’t released as part of PaaS SIG any origin rpm for v3.10.rc0 is because we couldn’t build it due to missing code which was made in OCP but has been ported to Origin but that is now sorted
Once that was done i was waiting for the new tag in Origin repo to be created which was done few hours ago with the official v3.10 release and so i’m working on it to get the rpms out for testing.
Any news here. Openshift 3.10 has been released by RedHat, however origin is not available
@DanyC97 prerequisites were not running in my environment due an error in one of my scripts. deploy_cluster.yaml was being running instead (yesterday). i fixed that.
i ran fresh again
$ git describeopenshift-ansible-3.10.29-1-22-gdec10a6d6i didn’t remove
atomicfromdefault_r_openshift_node_image_prep_packagesthis time, nor patchedroles/openshift_repos/tasks/centos_repos.ymlduring prerequisites execution:
$ cat /etc/yum.repos.d/CentOS-OpenShift-Origin.repo # master-01output
prerequisites ran to completion without errors with this repo file. So the PR in mention was actually ok!
running deploy_cluster.yaml now.
will re-open if another rpm issue is found.
thanks!
@sdodson just getting a couple of non related errors to this issue.
Hosts: prod-lb-01.example.com Play: Initialize cluster facts Task: Gather Cluster facts Message: MODULE FAILURE
Hosts: prod-master-01.example.com Play: Create Hosted Resources - router Task: Get the certificate contents for router Message: |lib_utils_oo_select_keys failed expects to filter on a dict or object
I’ll work in a solution, retry and let you know.
@DanyC97
sure, no matter i’ve set
openshift_release=v3.10andopenshift_repos_enable_testing=truein the inventory, the/etc/yum.repos.d/CentOS-OpenShift-Origin310.repofile does not appear in the server during prerequisites nor cluster_deploy steps, thence there’s no valid source to pull the rpms from. This happen in the current v3.10 branch ($ git describe => openshift-ansible-3.10.29-1-2-g8e9cd4a1b) which already has the PR changes merged.I was able to successfully test this part before, i think in 3.10-27-1, by locally patching roles/openshift_repos/tasks/centos_repos.yml and adding the roles/openshift_repos/templates/CentOS-OpenShift-Origin310.repo.j2 file. With the patch, ansible did copy the CentOS-OpenShift-Origin310.repo to the servers. see links with its respective content.
Maybe here there’s a clue: I noticed centos_repos.yml is no longer part of the PR nor the merge, which initially included this change:
CentOS-OpenShift-Origin{{ ((openshift_release | default('')).split('.') | join(''))[0:3] }}.repo.j2, but even if i patch the current branch with that sentence i cannot get the CentOS-OpenShift-Origin310.repo generated in the servers either. so that’s the reason i decided to manually copy the CentOS-OpenShift-Origin310.repo file to the servers before the test.hope this help to clarify.
@sdodson I haven’t tried that option yet but could in a next run.
i already did what @joshainglis suggested, to comment/remove “atomic” from default_r_openshift_node_image_prep_packages.
so far so good, this approach let the prerequisites continue to further steps.
I’ll keep you posted. waiting for results yet.
List of files under /etc/yum.repos.d/
Following is output of all files
@DanyC97 i patched my local openshift-ansible repo with the changes that you performed in the PR. I also removed any origin’s .repo file from the servers (fresh machines setup).
I can tell the new code worked! i.e. the right /etc/yum.repos.d/CentOS-OpenShift-Origin310.repo file was created and pre-requisites ran correctly.
looking forward they become available in the branch after approval/tests/commit.
now I’m running deploy_cluster.yaml.
so far so good. I’ll keep you posted of the final results on this part.
@vital-st1x please comment directly on the PR and not on this issue.
Trying to keep the history of all the comments close to the initial problem of missing rpms. I’ll look at your comments and respond on the PR itself, thanks !
oh !! thank you for this vital info, i’ll fire a PR and fix this too in addition to what i mentioned in my previous comment
update - PR created
My pleasure @DanyC97
This is another one when running deploy_cluster.yml
{ “attempts”: 3, “changed”: false, “invocation”: { “module_args”: { “allow_downgrade”: false, “bugfix”: false, “conf_file”: null, “disable_gpg_check”: false, “disable_plugin”: [], “disablerepo”: null, “enable_plugin”: [], “enablerepo”: null, “exclude”: null, “install_repoquery”: true, “installroot”: “/”, “list”: null, “name”: [ “atomic” ], “security”: false, “skip_broken”: false, “state”: “present”, “update_cache”: false, “update_only”: false, “validate_certs”: true } }, “item”: “atomic”, “msg”: “Error: Package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64 (extras)\n Requires: python-docker-py >= 1.7.2-1\n Available: docker-python-1.4.0-115.el7.x86_64 (extras)\n python-docker-py = 1.4.0-115.el7\n Available: python-docker-py-1.10.6-3.el7.noarch (extras)\n python-docker-py = 1.10.6-3.el7\n Available: python-docker-py-1.10.6-4.el7.noarch (extras)\n python-docker-py = 1.10.6-4.el7\n”, “rc”: 1, “results”: [ “Loaded plugins: fastestmirror\nLoading mirror speeds from cached hostfile\n * base: centos.mirror.lstn.net\n * extras: mirror.hostduplex.com\n * updates: centos.mirror.lstn.net\nResolving Dependencies\n–> Running transaction check\n—> Package atomic.x86_64 1:1.22.1-22.git5a342e3.el7 will be installed\n–> Processing Dependency: python-docker-py >= 1.7.2-1 for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\nPackage python-docker-py-1.10.6-4.el7.noarch is obsoleted by python2-docker-3.3.0-1.el7.noarch which is already installed\n–> Processing Dependency: runc for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\n–> Processing Dependency: python-dateutil for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\n–> Processing Dependency: gomtree for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\n–> Running transaction check\n—> Package atomic.x86_64 1:1.22.1-22.git5a342e3.el7 will be installed\n–> Processing Dependency: python-docker-py >= 1.7.2-1 for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\nPackage python-docker-py-1.10.6-4.el7.noarch is obsoleted by python2-docker-3.3.0-1.el7.noarch which is already installed\n—> Package gomtree.x86_64 0:0.5.0-0.2.git16da0f8.el7 will be installed\n—> Package python-dateutil.noarch 0:1.5-7.el7 will be installed\n—> Package runc.x86_64 0:1.0.0-27.rc5.dev.git4bb1fe4.el7 will be installed\n–> Processing Dependency: criu for package: runc-1.0.0-27.rc5.dev.git4bb1fe4.el7.x86_64\n–> Running transaction check\n—> Package atomic.x86_64 1:1.22.1-22.git5a342e3.el7 will be installed\n–> Processing Dependency: python-docker-py >= 1.7.2-1 for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\nPackage python-docker-py-1.10.6-4.el7.noarch is obsoleted by python2-docker-3.3.0-1.el7.noarch which is already installed\n—> Package criu.x86_64 0:3.5-4.el7 will be installed\n–> Processing Dependency: libprotobuf-c.so.1(LIBPROTOBUF_C_1.0.0)(64bit) for package: criu-3.5-4.el7.x86_64\n–> Processing Dependency: libprotobuf-c.so.1()(64bit) for package: criu-3.5-4.el7.x86_64\n–> Processing Dependency: libnet.so.1()(64bit) for package: criu-3.5-4.el7.x86_64\n–> Running transaction check\n—> Package atomic.x86_64 1:1.22.1-22.git5a342e3.el7 will be installed\n–> Processing Dependency: python-docker-py >= 1.7.2-1 for package: 1:atomic-1.22.1-22.git5a342e3.el7.x86_64\nPackage python-docker-py-1.10.6-4.el7.noarch is obsoleted by python2-docker-3.3.0-1.el7.noarch which is already installed\n—> Package libnet.x86_64 0:1.1.6-7.el7 will be installed\n—> Package protobuf-c.x86_64 0:1.0.2-3.el7 will be installed\n–> Finished Dependency Resolution\n You could try using --skip-broken to work around the problem\n You could try running: rpm -Va --nofiles --nodigest\n” ] }
prerequisites passed. running playbooks/deploy_cluster.yml now.
@imranrazakhan : add the testing repo
@DanyC97 Repo 3.10 for soon or not yet? Thank’s
@ticapix Is there any official repo for 3.10 rc?
@gbenhaim : any update or way we could help ?
@DanyC97 thanks.