9:42 p.m.
Simone, ovirt-users,
I'm running into an issue with the ovirt.hosted_engine_setup role. I can't
decide if my problem is 1) that the SSH connection to the HostedEngineLocal
VM repeatedly fails due to that new VM not having an entry in
~/.ssh/known_hosts or 2) that I don't have some variable defined
correctly. I can force my way past #1 by setting host_key_checking=false
in ansible.cfg, but if I do that, I immediately get an error about an
undefined hostvar variable... Can you point me in the right direction to
debug this? I would really like to be able to rebuild my RHHI/RHV cluster
using these gluster+ovirt roles.
Here is the undefined variable error, and the variable files I'm including
via the "ansible-playbook -e..." syntax:
TASK [ovirt.hosted_engine_setup : Wait for the local VM]
*****************************************************************************************
ok: [localhost -> rhvm.home.lab]
TASK [ovirt.hosted_engine_setup : Add an entry for this host on /etc/hosts
on the local VM] ******************************************************
fatal: [localhost]: FAILED! => {"msg": "The task includes an option with
an
undefined variable. *The error was: u\"hostvars['rhhi1.home.lab']\" is
undefined*\n\nThe error appears to be in
'/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml':
line 8, column 5, but may\nbe elsewhere in the file depending on the exact
syntax problem.\n\nThe offending line appears to be:\n\n timeout:
180\n - name: Add an entry for this host on /etc/hosts on the local VM\n
^ here\n"}
Thank you,
John Call
Red Hat - Storage Domain Architect
jcall(a)redhat.com
(714) 267-8802
On Tue, Jan 7, 2020 at 12:27 AM Gobinda Das <godas(a)redhat.com> wrote:
Hi John,
Sorry misunderstood, I thought you are using cockpit based deployment.
Your gluster_inventory is correct for end to end deployment.
On Tue, Jan 7, 2020 at 12:43 PM Gobinda Das <godas(a)redhat.com> wrote:
> Hi John,
> I am seeing error:
> TASK [ovirt.hosted_engine_setup : Get local VM dir path]
>
*****************************************************************************************
> task path:
>
/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml:57
> fatal: [localhost]: FAILED! => {
> "msg": "The task includes an option with an undefined variable.
The
> error was: 'ansible.vars.hostvars.HostVarsVars object' has no attribute
> 'he_local_vm_dir'\n\nThe error appears to be in
>
'/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml':
> line 57, column 7, but may\nbe elsewhere in the file depending on the exact
> syntax problem.\n\nThe offending line appears to be:\n\n delegate_to:
> \"{{ he_ansible_host_name }}\"\n - name: Get local VM dir path\n ^
> here\n"
> }
> Adding @Simone Tiraboschi <stirabos(a)redhat.com>
> Simone do we need to explicitly specify he_local_vm_dir ?
> In gluster_inventory you can't mention other hosts and SD informations to
> auto add. For that you need to create separate file under:
> /usr/share/ovirt-hosted-engine-setup/gdeploy-inventory.yml
> and add entries there like:
>
> gluster:
> hosts:
> host2:
> host3:
> vars:
> storage_domains:
>
[{"name":"data","host":"host1","address":"host1","path":"/data","mount_options":"backup-volfile-servers=host2:host3"},{"name":"vmstore","host":"host1","address":"host1","path":"/vmstore","mount_options":"backup-volfile-servers=host2:host3"}]
>
> Then hook will automatically read this file and perform action
> accordingly.
>
> On Tue, Jan 7, 2020 at 11:53 AM John Call <jcall(a)redhat.com> wrote:
>
>> Hi Gobinda, ovirt-users,
>>
>> I think I'm close, but I'm running into a timeout/error when the
>> ovirt.hosted-engine-setup tasks try to connect to the new
>> "HostedEngineLocal" VM. It seems to complain about host_key_checking.
I
>> disabled that via ansible.cfg, but it fails on the next task (see short
>> logs.) Do you see anything obviously wrong in my inventory and json
>> files? I've attached the verbose playbook logs when I run this from the
>> first HCI host...
>>
>> [root@rhhi1 hc-ansible-deployment]# pwd
>> /etc/ansible/roles/gluster.ansible/playbooks/hc-ansible-deployment
>>
>> [root@rhhi1 hc-ansible-deployment]# ansible-playbook -b -vvv -i
>> gluster_inventory.yml hc_deployment.yml
>> --extra-vars='(a)he_gluster_vars.json' | tee playbook.logs
>>
>>
>> Thank you,
>> John Call
>> Red Hat - Storage Domain Architect
>> jcall(a)redhat.com
>> (714) 267-8802
>>
>>
>> On Wed, Dec 11, 2019 at 11:09 PM Gobinda Das <godas(a)redhat.com> wrote:
>>
>>> Hi John,
>>> You need to specify storage-fqdn(Which should mapped to storage
>>> network) and ovirtmgmt-fqdn (Which should mapped to frontend network) like
>>> this:
>>> hc_nodes:
>>> hosts:
>>> host1-STORAGE-fqdn:
>>> host2-STORAGE-fqdn:
>>> host3-STORAGE-fqdn:
>>> vars:
>>> cluster_nodes:
>>> - host1-STORAGE-fqdn
>>> - host2-STORAGE-fqdn
>>> - host3-STORAGE-fqdn
>>> gluster_features_hci_cluster: "{{ cluster_nodes }}"
>>>
>>> gluster:
>>> host2-ovirtmgmt-fqdn:
>>> host3-ovirtmgmt-fqdn:
>>> storage_domains:
>>>
[{"name":"data","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/data","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"},{"name":"vmstore","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/vmstore","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"}]
>>>
>>>
>>> On Thu, Dec 12, 2019 at 2:47 AM John Call <jcall(a)redhat.com> wrote:
>>>
>>>> Hi ovirt-users,
>>>>
>>>> I'm trying to automate my HCI deployment, but can't figure out
how to
>>>> specify multiple network interfaces in gluster_inventory.yml. My
servers
>>>> have two NICs, one for ovirtmgmt (and everything else), and the other is
>>>> just for Gluster. How should I populate the inventory/vars file? Is
this
>>>> correct?
>>>>
>>>> [root@rhhi1 hc-ansible-deployment]# pwd
>>>> /etc/ansible/roles/gluster.ansible/playbooks/hc-ansible-deployment
>>>>
>>>> [root@rhhi1 hc-ansible-deployment]# cat gluster_inventory.yml
>>>> --lots of stuff omitted--
>>>> hc_nodes:
>>>> hosts:
>>>> host1-STORAGE-fqdn:
>>>> host2-STORAGE-fqdn:
>>>> host3-STORAGE-fqdn:
>>>> vars:
>>>> cluster_nodes:
>>>> - host1-ovirtmgmt-fqdn
>>>> - host2-ovirtmgmt-fqdn
>>>> - host3-ovirtmgmt-fqdn
>>>> gluster_features_hci_cluster: "{{ cluster_nodes }}"
>>>>
>>>> gluster:
>>>> host2-STORAGE-fqdn:
>>>> host3-STORAGE-fqdn:
>>>> storage_domains:
>>>>
[{"name":"data","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/data","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"},{"name":"vmstore","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/vmstore","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"}]
>>>> _______________________________________________
>>>> Users mailing list -- users(a)ovirt.org
>>>> To unsubscribe send an email to users-leave(a)ovirt.org
>>>> Privacy Statement: https://www.ovirt.org/site/privacy-policy/
>>>> oVirt Code of Conduct:
>>>> https://www.ovirt.org/community/about/community-guidelines/
>>>> List Archives:
>>>>
https://lists.ovirt.org/archives/list/users@ovirt.org/message/TEAHUX6LSPM...
>>>>
>>>
>>>
>>> --
>>>
>>>
>>> Thanks,
>>> Gobinda
>>>
>>
>
> --
>
>
> Thanks,
> Gobinda
>
--
Thanks,
Gobinda
8:53 p.m.
New subject: Question about HCI gluster_inventory.yml
Adding rhev-tech
Something I found interesting from the -vvvv ansible logs was the
difference in connect options between SSH and SFTP. I see the SSH connect
option "-o StrictHostKeyChecking=no" that was added to the new host
<https://github.com/oVirt/ovirt-ansible-hosted-engine-setup/blob/master/ta...;,
but I don't see those options on the subsequent SFTP command. I'm assuming
the SFTP command is uploading the ping module (maybe because of the SSH
proxy command logic linked above...) It seems to me that the logic of the
playbook link above is wrong. There is a conditional check that determines
whether or not the new HE will be subjected to key checking. Because this
is a brand new HE VM, nobody will know the host_key, and so
host_key_checking must not be done.
A few select lines from the ansible -vvvv logs:
<rhvm.home.lab> SSH: EXEC sshpass -d12 *ssh* -vvv -C -o ControlMaster=auto
-o ControlPersist=60s -o 'User="root"' -o ConnectTimeout=10 *-o
StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null* -o
'ProxyCommand=ssh -W %h:%p -q root(a)rhhi1.home.lab' -o
ControlPath=/root/.ansible/cp/9de9431b6b rhvm.home.lab '/bin/sh -c
'"'"'echo ~root && sleep 0'"'"''
...
wait_for_connection: attempting ping module test
Using module file
/usr/lib/python2.7/site-packages/ansible/modules/system/ping.py
<rhvm.home.lab> PUT /root/.ansible/tmp/ansible-local-31228wLAFQf/tmpnNJbIE
TO
/root/.ansible/tmp/ansible-tmp-1578434044.88-51347511337777/AnsiballZ_ping.py
<rhvm.home.lab> SSH: EXEC sshpass -d258 *sftp* -o BatchMode=no -b - -vvv -C
-o ControlMaster=auto -o ControlPersist=60s -o 'User="root"' -o
ConnectTimeout=10 -o ControlPath=/root/.ansible/cp/9de9431b6b
'[rhvm.home.lab]'
Thank you,
John Call
Red Hat - Storage Domain Architect
jcall(a)redhat.com
(714) 267-8802
On Tue, Jan 7, 2020 at 11:42 AM John Call <jcall(a)redhat.com> wrote:
Simone, ovirt-users,
I'm running into an issue with the ovirt.hosted_engine_setup role. I
can't decide if my problem is 1) that the SSH connection to the
HostedEngineLocal VM repeatedly fails due to that new VM not having an
entry in ~/.ssh/known_hosts or 2) that I don't have some variable defined
correctly. I can force my way past #1 by setting host_key_checking=false
in ansible.cfg, but if I do that, I immediately get an error about an
undefined hostvar variable... Can you point me in the right direction to
debug this? I would really like to be able to rebuild my RHHI/RHV cluster
using these gluster+ovirt roles.
Here is the undefined variable error, and the variable files I'm including
via the "ansible-playbook -e..." syntax:
TASK [ovirt.hosted_engine_setup : Wait for the local VM]
*****************************************************************************************
ok: [localhost -> rhvm.home.lab]
TASK [ovirt.hosted_engine_setup : Add an entry for this host on /etc/hosts
on the local VM] ******************************************************
fatal: [localhost]: FAILED! => {"msg": "The task includes an option
with
an undefined variable. *The error was: u\"hostvars['rhhi1.home.lab']\"
is
undefined*\n\nThe error appears to be in
'/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml':
line 8, column 5, but may\nbe elsewhere in the file depending on the exact
syntax problem.\n\nThe offending line appears to be:\n\n timeout:
180\n - name: Add an entry for this host on /etc/hosts on the local VM\n
^ here\n"}
Thank you,
John Call
Red Hat - Storage Domain Architect
jcall(a)redhat.com
(714) 267-8802
On Tue, Jan 7, 2020 at 12:27 AM Gobinda Das <godas(a)redhat.com> wrote:
> Hi John,
> Sorry misunderstood, I thought you are using cockpit based deployment.
> Your gluster_inventory is correct for end to end deployment.
>
> On Tue, Jan 7, 2020 at 12:43 PM Gobinda Das <godas(a)redhat.com> wrote:
>
>> Hi John,
>> I am seeing error:
>> TASK [ovirt.hosted_engine_setup : Get local VM dir path]
>>
*****************************************************************************************
>> task path:
>>
/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml:57
>> fatal: [localhost]: FAILED! => {
>> "msg": "The task includes an option with an undefined
variable. The
>> error was: 'ansible.vars.hostvars.HostVarsVars object' has no attribute
>> 'he_local_vm_dir'\n\nThe error appears to be in
>>
'/usr/share/ansible/roles/ovirt.hosted_engine_setup/tasks/bootstrap_local_vm/03_engine_initial_tasks.yml':
>> line 57, column 7, but may\nbe elsewhere in the file depending on the exact
>> syntax problem.\n\nThe offending line appears to be:\n\n delegate_to:
>> \"{{ he_ansible_host_name }}\"\n - name: Get local VM dir path\n
^
>> here\n"
>> }
>> Adding @Simone Tiraboschi <stirabos(a)redhat.com>
>> Simone do we need to explicitly specify he_local_vm_dir ?
>> In gluster_inventory you can't mention other hosts and SD informations
>> to auto add. For that you need to create separate file under:
>> /usr/share/ovirt-hosted-engine-setup/gdeploy-inventory.yml
>> and add entries there like:
>>
>> gluster:
>> hosts:
>> host2:
>> host3:
>> vars:
>> storage_domains:
>>
[{"name":"data","host":"host1","address":"host1","path":"/data","mount_options":"backup-volfile-servers=host2:host3"},{"name":"vmstore","host":"host1","address":"host1","path":"/vmstore","mount_options":"backup-volfile-servers=host2:host3"}]
>>
>> Then hook will automatically read this file and perform action
>> accordingly.
>>
>> On Tue, Jan 7, 2020 at 11:53 AM John Call <jcall(a)redhat.com> wrote:
>>
>>> Hi Gobinda, ovirt-users,
>>>
>>> I think I'm close, but I'm running into a timeout/error when the
>>> ovirt.hosted-engine-setup tasks try to connect to the new
>>> "HostedEngineLocal" VM. It seems to complain about
host_key_checking. I
>>> disabled that via ansible.cfg, but it fails on the next task (see short
>>> logs.) Do you see anything obviously wrong in my inventory and json
>>> files? I've attached the verbose playbook logs when I run this from the
>>> first HCI host...
>>>
>>> [root@rhhi1 hc-ansible-deployment]# pwd
>>> /etc/ansible/roles/gluster.ansible/playbooks/hc-ansible-deployment
>>>
>>> [root@rhhi1 hc-ansible-deployment]# ansible-playbook -b -vvv -i
>>> gluster_inventory.yml hc_deployment.yml
>>> --extra-vars='(a)he_gluster_vars.json' | tee playbook.logs
>>>
>>>
>>> Thank you,
>>> John Call
>>> Red Hat - Storage Domain Architect
>>> jcall(a)redhat.com
>>> (714) 267-8802
>>>
>>>
>>> On Wed, Dec 11, 2019 at 11:09 PM Gobinda Das <godas(a)redhat.com> wrote:
>>>
>>>> Hi John,
>>>> You need to specify storage-fqdn(Which should mapped to storage
>>>> network) and ovirtmgmt-fqdn (Which should mapped to frontend network)
like
>>>> this:
>>>> hc_nodes:
>>>> hosts:
>>>> host1-STORAGE-fqdn:
>>>> host2-STORAGE-fqdn:
>>>> host3-STORAGE-fqdn:
>>>> vars:
>>>> cluster_nodes:
>>>> - host1-STORAGE-fqdn
>>>> - host2-STORAGE-fqdn
>>>> - host3-STORAGE-fqdn
>>>> gluster_features_hci_cluster: "{{ cluster_nodes }}"
>>>>
>>>> gluster:
>>>> host2-ovirtmgmt-fqdn:
>>>> host3-ovirtmgmt-fqdn:
>>>> storage_domains:
>>>>
[{"name":"data","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/data","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"},{"name":"vmstore","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/vmstore","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"}]
>>>>
>>>>
>>>> On Thu, Dec 12, 2019 at 2:47 AM John Call <jcall(a)redhat.com>
wrote:
>>>>
>>>>> Hi ovirt-users,
>>>>>
>>>>> I'm trying to automate my HCI deployment, but can't figure
out how to
>>>>> specify multiple network interfaces in gluster_inventory.yml. My
servers
>>>>> have two NICs, one for ovirtmgmt (and everything else), and the other
is
>>>>> just for Gluster. How should I populate the inventory/vars file? Is
this
>>>>> correct?
>>>>>
>>>>> [root@rhhi1 hc-ansible-deployment]# pwd
>>>>> /etc/ansible/roles/gluster.ansible/playbooks/hc-ansible-deployment
>>>>>
>>>>> [root@rhhi1 hc-ansible-deployment]# cat gluster_inventory.yml
>>>>> --lots of stuff omitted--
>>>>> hc_nodes:
>>>>> hosts:
>>>>> host1-STORAGE-fqdn:
>>>>> host2-STORAGE-fqdn:
>>>>> host3-STORAGE-fqdn:
>>>>> vars:
>>>>> cluster_nodes:
>>>>> - host1-ovirtmgmt-fqdn
>>>>> - host2-ovirtmgmt-fqdn
>>>>> - host3-ovirtmgmt-fqdn
>>>>> gluster_features_hci_cluster: "{{ cluster_nodes }}"
>>>>>
>>>>> gluster:
>>>>> host2-STORAGE-fqdn:
>>>>> host3-STORAGE-fqdn:
>>>>> storage_domains:
>>>>>
[{"name":"data","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/data","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"},{"name":"vmstore","host":"host1-STORAGE-fqdn","address":"host1-STORAGE-fqdn","path":"/vmstore","mount_options":"backup-volfile-servers=host2-STORAGE-fqdn:host3-STORAGE-fqdn"}]
>>>>> _______________________________________________
>>>>> Users mailing list -- users(a)ovirt.org
>>>>> To unsubscribe send an email to users-leave(a)ovirt.org
>>>>> Privacy Statement: https://www.ovirt.org/site/privacy-policy/
>>>>> oVirt Code of Conduct:
>>>>> https://www.ovirt.org/community/about/community-guidelines/
>>>>> List Archives:
>>>>>
https://lists.ovirt.org/archives/list/users@ovirt.org/message/TEAHUX6LSPM...
>>>>>
>>>>
>>>>
>>>> --
>>>>
>>>>
>>>> Thanks,
>>>> Gobinda
>>>>
>>>
>>
>> --
>>
>>
>> Thanks,
>> Gobinda
>>
>
>
> --
>
>
> Thanks,
> Gobinda
>
1780
days inactive
1781
days old
1 comments
1 participants
participants (1)
-
John Call