Red Hat Bugzilla – Bug 1465377
[RFE] Recovery of cns from Openshift backup and restore option
Last modified: 2017-11-15 07:18:59 EST
Description of problem: The procedure consists of the following steps:
0. Stop ocp services (api, controllers, node)
1. Restore etcd data
2. Restore ocp master config
3. Redeploy certificates
5. Start up ocp services (api, controllers, node)
6. delete nodes to get current machine id (cloudprovider)
7. relabel nodes (master -> unschedulable, gluster -> storagenode=glusterfs)
8. reboot nodes
Note: old directories of gluster and service accounts are in place. After this point deployments, builds, scaling, dynamically provision cinder storage and the sdn works.
Gluster is the only part which does not come back as intended i.e gluster pod are in running state but heketi pod is continuously going in crash loop state
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Actual results: It is not in running state.
Expected results: It should come up after restoration.