Red Hat Bugzilla – Bug 1304954
[openshift-doc] Need documentation of adding new node by scale up playbooks
Last modified: 2016-07-15 13:22:20 EDT
Description of problem: - From v3.1.x, playbook scaleup.yml has been released and we can use scale up playbook to add new node. It will not make any change to the existing node. Version-Release number of selected component (if applicable): - v3.1 Document URL: - There are no existing doc. - Maybe it is beeter to add it to this https://docs.openshift.com/enterprise/3.1/admin_guide/manage_nodes.html Additional info: - There is a KCS https://access.redhat.com/solutions/2150381
Working on this this week.
Hi Alex, Thanks for working on that. I'm quite curious about the recommended/rigt way to use this scaleup script. Is this task of documenting it still on track or has it been postponed? Thanks anyways
(In reply to fred from comment #4) > Hi Alex, > > Thanks for working on that. I'm quite curious about the recommended/rigt way > to use this scaleup script. Is this task of documenting it still on track or > has it been postponed? > > Thanks anyways Sorry, it got waylaid since my last comment, but definitely still high on my list. I hope to have something within the week. Apologies for the delay.
Sorry, finally getting back to this. See the description in the following PR for links to internal preview builds of what I've got so far: https://github.com/openshift/openshift-docs/pull/1875
Commit pushed to master at https://github.com/openshift/openshift-docs https://github.com/openshift/openshift-docs/commit/7ef3c8e5b70dceed8a3c6e5b5515b7ccf17c6fbc Merge pull request #1875 from adellape/add_node Bug 1304954: Updated steps for adding nodes
PR tech/peer reviewed and merged. Pending release in the next OSE 3.1 docs update (next week).
I want to note that the scaleup playbook seems to run oc commands as the OpenShift user currently logged in. So we must inform users to log in as system:admin before running the playbook, otherwise it may fail at labeling nodes with error: stderr: Error from server: User "myuser" cannot update nodes at the cluster scope
quick update - just tried to reproduce it with a non-default user with cluster admin rights and it worked. So probably just related to permissions, will re-test later with a limited user.
I logged bz#1327409 regarding this behavior. In general - if the user logged into oc does not have cluster-admin permissions, the playbook will fail.
This doc* worked fine for v3.1 does it work on v3.2? I assume it will still work, but the doc is titled: "How to add new Node in OpenShift v3.1" * https://access.redhat.com/solutions/2150381
(In reply to dlbewley from comment #12) > This doc* worked fine for v3.1 does it work on v3.2? > > I assume it will still work, but the doc is titled: > > "How to add new Node in OpenShift v3.1" > > * https://access.redhat.com/solutions/2150381 The file path changed slightly in 3.2 (noted in step 2 of that KBase). The OSE 3.2 docs have steps for adding nodes for both the advanced and quick install paths: https://docs.openshift.com/enterprise/3.2/install_config/install/advanced_install.html#adding-nodes-advanced https://docs.openshift.com/enterprise/3.2/install_config/install/quick_install.html#quick-before-you-begin