On Sat, Jan 9, 2021 at 7:49 PM Alexey Nikolaev
<alexeynikolaev.post(a)yandex.ru> wrote:
Hi community!
I have issue when trying to restore 4.3.10.4 hosted engine.
[ ERROR ] Exception: Not possible to manage storage domain 'hosted_storage'.
[ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg":
"Not possible to manage storage domain 'hosted_storage'."}
This error occurs with both Glusterfs and NFS shared storage.
engine setup ansible create storage domain log
https://gist.github.com/virtio-technique/38a3678b834017f3e4cd9c03f6811e71
vdsm log
https://gist.github.com/virtio-technique/8c5ad1ce0e2f07d98d61a964744adff9
hosted engine setup log
https://gist.github.com/virtio-technique/00230b117cf8ee1b1a094c1001f424bd
It seems like you provided storage creds, then setup tried to do stuff
there, then:
2021-01-09 17:01:42,949+0300 DEBUG
otopi.ovirt_hosted_engine_setup.ansible_utils
ansible_utils._process_output:103 {u'invocation': {u'module_args':
{u'comment': None, u'warning_low_space': None, u'glusterfs':
None,
u'localfs': None, u'managed_block_storage': None, u'data_center':
u'Default', u'id': None, u'iscsi': None, u'state':
u'present',
u'wipe_after_delete': None, u'destroy': None, u'fcp': None,
u'description': None, u'format': None, u'nested_attributes': [],
u'host': u'node77-05.aaa.fsin.uis', u'discard_after_delete':
None,
u'wait': True, u'domain_function': u'data', u'name':
u'hosted_storage', u'critical_space_action_blocker': None,
u'posixfs':
None, u'poll_interval': 3, u'fetch_nested': False, u'nfs': None,
u'timeout': 180, u'backup': None}}, u'msg': u'Timeout exceed
while
waiting on result state of the entity.', u'exception': u'Traceback
(most recent call last):\n File
"/tmp/ansible_ovirt_storage_domain_payload_tgcLDH/ansible_ovirt_storage_domain_payload.zip/ansible/modules/cloud/ovirt/ovirt_storage_domain.py",
line 771, in main\n File
"/tmp/ansible_ovirt_storage_domain_payload_tgcLDH/ansible_ovirt_storage_domain_payload.zip/ansible/modules/cloud/ovirt/ovirt_storage_domain.py",
line 635, in post_create_check\n File
"/tmp/ansible_ovirt_storage_domain_payload_tgcLDH/ansible_ovirt_storage_domain_payload.zip/ansible/module_utils/ovirt.py",
line 364, in wait\n raise Exception("Timeout exceed while waiting
on result state of the entity.")\nException: Timeout exceed while
waiting on result state of the entity.\n', u'changed': False,
u'_ansible_no_log': False}
2021-01-09 17:01:43,050+0300 ERROR
otopi.ovirt_hosted_engine_setup.ansible_utils
ansible_utils._process_output:107 Exception: Timeout exceed while
waiting on result state of the entity.
and then you provided again the same creds and it failed with the
message you quoted.
I suggest to clean up this space and only then try again.
If it fails again due to "Timeout exceed", try to understand why, and fix this.
If it's simply due to stuff being slow (with no good reason or
anything you can affect), you can try making the timeout longer, by
patching create_storage_domain.yml, adding a "timeout" parameter (in
seconds), like this:
- name: Activate storage domain
ovirt_storage_domain:
host: "{{ he_host_name }}"
data_center: "{{ datacenter_name }}"
name: "{{ he_storage_domain_name }}"
wait: true
state: present
auth: "{{ ovirt_auth }}"
timeout: 3600
when: storage_domain_details.ovirt_storage_domains[0].available|int
= required_size|int
register: otopi_storage_domain_details
In 4.3, this file is in
/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/create_storage_domain.yml
.
In current 4.4, it's in
/usr/share/ansible/collections/ansible_collections/ovirt/ovirt/roles/hosted_engine_setup/tasks/create_storage_domain.yml
.
Can I use a hosted engine backup file to restore the manager to a
dedicated server and manually delete the hosted engine domain?
Please search the list archives, e.g.:
https://www.mail-archive.com/users@ovirt.org/msg60933.html
Good luck and best regards,
--
Didi