Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1592672

Summary: docker-registry not deploying because installer failing to label nodes with region in 3.10.1-1
Product: OpenShift Container Platform Reporter: Nicholas Schuetz <nick>
Component: InstallerAssignee: Scott Dodson <sdodson>
Status: CLOSED NOTABUG QA Contact: Johnny Liu <jialiu>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 3.10.0CC: aos-bugs, jokerman, mmccomas, nick, vrutkovs
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-06-19 15:26:50 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
ansible inventory
none
error none

Description Nicholas Schuetz 2018-06-19 06:20:48 UTC
Created attachment 1452819 [details]
ansible inventory

TASK [openshift_hosted : Wait for registry pods] ******************************************************************************************************************************************************************
...
FAILED - RETRYING: Wait for registry pods (1 retries left).
fatal: [master01.ocp.nicknach.net]: FAILED! => {"attempts": 60, "changed": false, "failed": true, "results": {"cmd": "/usr/bin/oc get pod --selector=docker-registry=default -o json -n default", "results": [{"apiVersion": "v1", "items": [], "kind": "List", "metadata": {"resourceVersion": "", "selfLink": ""}}], "returncode": 0}, "state": "list"}

To get a clean install, i have to catch it failing and add some labels in another terminal.

oc label node infra01.ocp.nicknach.net region=infra zone=default --overwrite
oc label node infra02.ocp.nicknach.net region=infra zone=default --overwrite
oc label node infra03.ocp.nicknach.net region=infra zone=default --overwrite
oc label node master01.ocp.nicknach.net region=masters zone=default --overwrite
oc label node master02.ocp.nicknach.net region=masters zone=default --overwrite
oc label node master03.ocp.nicknach.net region=masters zone=default --overwrite
oc label node node03.ocp.nicknach.net region=primary zone=default --overwrite
oc label node node02.ocp.nicknach.net region=primary zone=default --overwrite
oc label node node01.ocp.nicknach.net region=primary zone=default --overwrite

After i do this, the registry-deploy container deploys and the install completes.  

Screenshot ansible hosts file attached.

Note:  I am using the new openshift_node_group_name as well as the legacy openshift_node_labels.

Comment 1 Nicholas Schuetz 2018-06-19 06:21:36 UTC
Created attachment 1452820 [details]
error

Comment 2 Nicholas Schuetz 2018-06-19 06:25:47 UTC
Here is the event:
  

17m         32m          57        docker-registry-1-deploy.153979c62983aa2d    Pod                                                   Warning   FailedScheduling          default-scheduler                          0/9 nodes are available: 9 node(s) didn't match node selector.

I dont think 'openshift_node_labels' is working anymore.

-Nick

Comment 3 Vadim Rutkovsky 2018-06-19 09:38:00 UTC
openshift_node_group_name="node-config-infra" would set nodelabels 'node-role.kubernetes.io/infra=true', so the selector won't find nodes as it expects them to be 'region=infra'.

See https://trello.com/c/7m7A7Vpu/579-5-standardize-on-node-rolekubernetesio-mastercomputeinfratrue

Nick, does it work if openshift_hosted_registry_selector is updated?

Comment 4 Nicholas Schuetz 2018-06-19 15:13:18 UTC
Thanks!  Commenting out all my legacy node selectors (in my ansible inventory) got me a clean install.  I'll port those to the new style for a more permanent solution.

-Nick

Comment 5 Vadim Rutkovsky 2018-06-19 15:26:50 UTC
Awesome, closing this as NOTABUG