Re: [ceph-ansible] rolling-upgrade variables not present

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Gregory,

> $ time ansible-playbook infrastructure-playbooks/rolling_update.yml -e
ireallymeanit=yes

You mentioned you were using /etc/ansible/hosts as the ansible inventory
file. But where are located the group_vars and host_vars directories ?

It looks like you have your group_vars and host_vars directories located
inside the main ceph-ansible directory.
But when running the infrastructure-playbooks/rolling_update.yml playbook
then ansible loads the variables either
- from the playbook directory (infrastructure-playbooks here)
- from the ansible inventory directory (/etc/ansible in your case)

If I'm right, I would suggest you to not store your {group,host)_vars
directories inside the ceph-ansible sources.
A better solution would be to keep the variables and the inventory in the
same directory like /etc/ansible in your case or any
other directory you want to create (but be sure to reference the ansible
inventory with -i <path> with the ansible-playbook cli)

Regards,

Dimitri

On Thu, Aug 19, 2021 at 6:03 AM Gregory Orange <gregory.orange@xxxxxxxxxxxxx>
wrote:

> I deployed Ceph 14.2.16 with ceph-ansible stable-4.0 a while back, and
> want to test upgrading. So for now I am trying rolling_update.yml for
> latest 14.x (before trying stable-5.0 and 15.x) but getting some errors,
> which seem to indicate empty or missing variables.
>
> Initially monitor_interface was unavailable for mons even though I have
> it in group_vars/all.yml and it worked when I deployed it. So I put
> 'monitor_interface=storage' in /etc/ansible/hosts against each mon
> entry, and got past that. Now I get this ...
>
> $ time ansible-playbook infrastructure-playbooks/rolling_update.yml -e
> ireallymeanit=yes
> ...
> TASK [ceph-validate : validate ceph_origin]
>
> ***********************************************************************************************
> Thursday 19 August 2021  09:57:06 +0000 (0:00:00.148)       0:00:36.032
> *******
> fatal: [dev-ceph-mon-01]: FAILED! => changed=false
>    msg: ceph_origin must be either 'repository', 'distro' or 'local'
> ...
>
> ... which indicates ceph_origin is unavailable, also from
> group_vars/all.yml. The only thing I could find in archives here or the
> deprecated ceph-ansible list is this:
>
> http://lists.ceph.com/pipermail/ceph-ansible-ceph.com/2020-April/000342.html
>
> But that doesn't give me a solution. Can someone suggest what I should
> try next?
>
> Thanks,
> Greg.
> _______________________________________________
> ceph-users mailing list -- ceph-users@xxxxxxx
> To unsubscribe send an email to ceph-users-leave@xxxxxxx
>
>
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx



[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux