k8s-infra icon indicating copy to clipboard operation
k8s-infra copied to clipboard

Program the daily shudown of servers

Open jacobdotcosta opened this issue 5 years ago • 6 comments

Solutions to implement the daily shutdown of servers.

We could build an ansible playbook to apply all basic configuration to vms. On, or the first, being a CRON job that would poweroff servers every night.

jacobdotcosta avatar Feb 19 '20 10:02 jacobdotcosta

being a CRON job that would poweroff servers every night.

As k8s cluster offers such an option, we could just create a Scheduled Pod responsible to execute the playbook able to shutdown and re-start the VMs. This scheduled pod could be deployed on our permanent Hetzner VM.

WDYT ? @geoand @jacobdotcosta @metacosm

cmoulliard avatar Feb 19 '20 10:02 cmoulliard

The first I think I would look into is if Hetzner cloud offers such a capability. I think other clouds have this so it's probably worth exploring first before any custom solution is developed

geoand avatar Feb 19 '20 10:02 geoand

Doing it automatically only makes sense since we're pretty much all in the same time zone. I also agree that it's worth investigating whether we can do this automatically using Hetzner directly.

On Wed, Feb 19, 2020 at 11:37 AM Georgios Andrianakis < [email protected]> wrote:

The first I think I would look into is if Hetzner cloud offers such a capability. I think other clouds have this so it's probably worth exploring first before any custom solution is developed

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/snowdrop/k8s-infra/issues/140?email_source=notifications&email_token=AAA5J6KKYKD7JWXS2PJBVDTRDUDVBA5CNFSM4KXWLMEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMHIEVY#issuecomment-588153431, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAA5J6JC3KNUWS32QLCXGVDRDUDVBANCNFSM4KXWLMEA .

metacosm avatar Feb 19 '20 14:02 metacosm

Hetzner dont provide option to shutdown or start Cloud VMs

On Wed, Feb 19, 2020 at 3:31 PM Chris Laprun [email protected] wrote:

Doing it automatically only makes sense since we're pretty much all in the same time zone. I also agree that it's worth investigating whether we can do this automatically using Hetzner directly.

On Wed, Feb 19, 2020 at 11:37 AM Georgios Andrianakis < [email protected]> wrote:

The first I think I would look into is if Hetzner cloud offers such a capability. I think other clouds have this so it's probably worth exploring first before any custom solution is developed

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub < https://github.com/snowdrop/k8s-infra/issues/140?email_source=notifications&email_token=AAA5J6KKYKD7JWXS2PJBVDTRDUDVBA5CNFSM4KXWLMEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMHIEVY#issuecomment-588153431 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/AAA5J6JC3KNUWS32QLCXGVDRDUDVBANCNFSM4KXWLMEA

.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/snowdrop/k8s-infra/issues/140?email_source=notifications&email_token=AADRHLUG5URLZTTNIVTDMY3RDU7E5A5CNFSM4KXWLMEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMICZIA#issuecomment-588262560, or unsubscribe https://github.com/notifications/unsubscribe-auth/AADRHLTWFBOMPZ2JQO4CJ2LRDU7E5ANCNFSM4KXWLMEA .

-- Charles Moulliard Apache Committer & PMC / Architect @RedHat Twitter : @cmoulliard | Blog : http://cmoulliard.github.io

cmoulliard avatar Feb 19 '20 15:02 cmoulliard

being a CRON job that would poweroff servers every night.

As k8s cluster offers such an option, we could just create a Scheduled Pod responsible to execute the playbook able to shutdown and re-start the VMs. This scheduled pod could be deployed on our permanent Hetzner VM.

WDYT ? @geoand @jacobdotcosta @metacosm

It sounds nice, although I'm not sure how to do that. Should we have 2 pods one to shutdown and another one to start. I could build the Ansible playbooks that do these tasks.

jacobdotcosta avatar Feb 20 '20 11:02 jacobdotcosta

Should we have 2 pods one to shutdown and another one to start. I could build the Ansible playbooks that do these tasks.

One pod is enough and we will use a k8s cronjob -> https://kubernetes.io/docs/tasks/job/automated-tasks-with-cron-jobs/

cmoulliard avatar Feb 20 '20 13:02 cmoulliard