{"build_id": "2014-07-21_15-51-31", "ostf_sha": "9863db951a6e159f4fa6e6861c8331e1af069cf8", "build_number": "443", "auth_required": false, "api": "1.0", "nailgun_sha": "bdd0bdec2b45eea843d559b7648bd5dca4873c66", "production": "docker", "fuelmain_sha": "539a5bf7493a5d14690a34bb18c3ad1c75b4f37f", "astute_sha": "fd9b8e3b6f59b2727b1b037054f10e0dd7bd37f1", "feature_groups": ["mirantis"], "release": "5.1", "fuellib_sha": "aa6cccdeae74553378475773642df44709027a48"}
Steps to reproduce:
1. Deploy cluster: Centos Multinode, 1 Controller 1 Compute, nova Flat
2. Backup master node #dockerctl backup
3. Add one more Compute node
4. Restore master #dockerctl restore /path/to/arch
Expected result:
Or we need add to specification notes like if you have changed cluster configuration after backup it is necessary to make new backup or fix issue with cluster states after restore.
Actual result:
# fuel node
id | status | name | cluster | ip | mac | roles | pending_roles | online
---|----------|---------------------|---------|--------------|-------------------|------------|---------------|-------
3 | discover | Untitled (46:50) | None | 10.108.0.139 | 64:3c:69:bb:46:50 | | | True
2 | ready | slave-02_compute | 1 | 10.108.0.4 | 64:fd:8d:c7:6b:43 | compute | | True
1 | ready | slave-01_controller | 1 | 10.108.0.3 | 64:38:1e:ac:38:ea | controller | | True
# nova hypervisor-list
+----+--------------------------+
| ID | Hypervisor hostname |
+----+--------------------------+
| 1 | node-2.test.domain.local |
| 2 | node-3.test.domain.local |
+----+--------------------------+
This should be converted to a blueprint. What you need is the ability to make an incremental backup of your deployment state. It's not so difficult to do this and append to the previous backup, but it's a feature.