minimize downtime during new deployments

[LesnyRumcajs]

Summary of changes
Changes introduced in this pull request:

• this lifecycle option should minimize the downtime caused by re-deployments by first creating the new service and only then scraping the old one.

Reference issue to close (if applicable)

Closes

Other information and links

[github-actions]

Forest: Calibnet Infrastructure Plan: success

Show Plan

module.calibnet.data.digitalocean_project.forest_project: Reading...
module.calibnet.data.digitalocean_ssh_keys.keys: Reading...
module.calibnet.data.digitalocean_ssh_keys.keys: Read complete after 0s [id=ssh_keys/14512061520513425405]
module.calibnet.digitalocean_droplet.forest: Refreshing state... [id=379939055]
module.calibnet.digitalocean_firewall.forest_firewall: Refreshing state... [id=179cfbb1-b532-4d06-bce7-334ec185ea3d]
module.calibnet.data.digitalocean_project.forest_project: Read complete after 1s [id=da5e6601-7fd9-4d02-951e-390f7feb3411]
module.calibnet.digitalocean_project_resources.connect_forest_project: Refreshing state... [id=da5e6601-7fd9-4d02-951e-390f7feb3411]

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # module.calibnet.digitalocean_droplet.forest will be updated in-place
  ~ resource "digitalocean_droplet" "forest" {
        id                   = "379939055"
        name                 = "forest-calibnet"
        tags                 = [
            "calibnet",
        ]
      # Warning: this attribute value will be marked as sensitive and will not
      # display in UI output after applying this change. The value is unchanged.
      ~ user_data            = (sensitive value)
        # (23 unchanged attributes hidden)
    }

Plan: 0 to add, 1 to change, 0 to destroy.

─────────────────────────────────────────────────────────────────────────────

Saved the plan to: tfplan

To perform exactly these actions, run the following command to apply:
    terraform apply "tfplan"

[lemmih]

Create-before-destroy caused problems in the past with NewRelic. But I trust you can do the NewRelic configuration to properly deal with this.

[LesnyRumcajs]

Create-before-destroy caused problems in the past with NewRelic. But I trust you can do the NewRelic configuration to properly deal with this.

@lemmih What kind of problems?

[lemmih]

Create-before-destroy caused problems in the past with NewRelic. But I trust you can do the NewRelic configuration to properly deal with this.

@lemmih What kind of problems?

I don't know any of the details but there were problems with telling NewRelic which nodes were supposed to be up and which were not.

[LesnyRumcajs]

Alright, let me look into it more beforehand, then. Sounds fishy.

[ruseinov]

Looks ok, I'll approve. Curious about the newrelic issue.

[samuelarogbonlo]

Alright, let me look into it more beforehand, then. Sounds fishy.

it has been fixed, the problem was new-relic was monitoring all nodes with the agent installed not the ones that was supposed to be online, since then we have added the optional installation of new relic and also filtered to specific hosts.