[DigitalOcean] Fix DO Tag issue

[srikiz]

We add tags to the droplets in order to match the correct volume that needs to be attached. The volumes for etcd-main and etcd-events are also tagged and the etcd-manager does this mapping.

Due to this change, this was breaking DO droplets to volume matching logic and the new clusters wasn't coming up.

FYI - @rifelpet @timoreimann

I can work on a better logic going further, but would need this change to go in so I can continue to test the e2e tests.

Please let me know your thoughts, thanks !

[rifelpet]

Ah, nice find! I believe we only prepend etcd- if the etcd member name starts with a digit. Should we add that same logic here as well?

cc @hakman

[hakman]

@rifelpet @srikiz We should keep the logic somehow, otherwise it will match only some of the cases.

[srikiz]

@hakman @rifelpet - In the case of DO, we don't actually have an availability zone. All the master droplets will be deployed in the same region. (example for torento region is tor1). So for a master count of 3, etcd-cluster names that gets into the trim function would be master-1-tor1-1, master-1-tor1-2, master-1-tor1-3. And the output would be "etcd-1", "etcd-2", "etcd-3". Previously the output was always "1" or "2" or "3" based on the number of masters.

For the test case examples as mentioned here it always results in the below output based on the above scenario. So I think it's very safe to assume the name as "etcd-1" or "etcd-2" etc. in DO case.

		{
			Input:  []string{"us-test-1a-1", "us-test-1a-2", "us-test-1a-3"},
			Output: []string{"etcd-1", "etcd-2", "etcd-3"},
		},

Please let me know if this helps..
FYI - @timoreimann

[hakman]

@srikiz Can it be a single master droplet? How does it look then?

[srikiz]

@hakman - for a single master cluster it looks like below.
`
etcdClusters:

• cpuRequest: 200m
  etcdMembers:
  • instanceGroup: master-tor1-1
    name: etcd-1
    memoryRequest: 100Mi
    name: main
• cpuRequest: 100m
  etcdMembers:
  • instanceGroup: master-tor1-1
    name: etcd-1
    memoryRequest: 100Mi
    name: events
    `

For a multi-master cluster, it looks as below -
`etcdClusters:

• cpuRequest: 200m
  etcdMembers:
  • instanceGroup: master-tor1-1
    name: etcd-1
  • instanceGroup: master-tor1-2
    name: etcd-2
  • instanceGroup: master-tor1-3
    name: etcd-3
    memoryRequest: 100Mi
    name: main
• cpuRequest: 100m
  etcdMembers:
  • instanceGroup: master-tor1-1
    name: etcd-1
  • instanceGroup: master-tor1-2
    name: etcd-2
  • instanceGroup: master-tor1-3
    name: etcd-3
    memoryRequest: 100Mi
    name: events`

[hakman]

Then I think the change should be ok. What do you think @rifelpet ?

[rifelpet]

agreed. lets give it a shot 👍🏻

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: rifelpet

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [rifelpet]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment