Description

All the ability for a user to add and edit the following fields for an HCP cluster machine pool:

• Node drain timeout (numerical value - minutes)
• maxunavailable (numerical value - number of nodes)
• maxsurge (numerical value - number of nodes

Acceptance criteria

• When adding a new machine pool to an existing HCP cluster, the user may (but is not required to) enter values for these fields
• When editing an existing machine pool to an existing HCP cluster the user may (but is not required to) modify these values
• When creating an new machine pool for any other cluster besides an AWS HCP cluster, the option to view/edit these fields will not be shown.
• When modifying an existing machine pools for any other cluster besides an AWS HCP cluster, the option to view/edit these fields will not be shown.
• The user is notified and is not allowed to continue if the user enters a non valid value.  Valid entries are:
  • max_surge: Greater than 0 (no upper limit)
  • max_unavailable: Greater than 0 (no upper limit)
  • node_drain_grace_period: Greater than or equal to 0 and less than or equal to 10080
• This change should not impact ROSA Classic and OSD clusters.

Mockups/Design

Mock-up document

Simplified design doc

Out of scope

Other changes including functionality and visibility of other fields (except for location) in the add/edit modal are out of scope. Existing behaviors, visibility rules, help text, and validation will remain unchanged.

Showing these values in list machine pool table - even in the expanded rows

Adding the values when creating an HCP cluster - this feature will be handled by another epic.

Testing implications

The recommended design will change the add/edit modal which will impact all clusters that allow the users to add/edit machine pools (ROSA classic, OSD, etc), so testing all of these cluster types will be required.

------------------------------------------------------------------------------------------------------------------------

Implementation notes

The ability to add/modify these values already exists in the API according to the doc:

The ability to view/add/edit the new fields for a node pool is available in the node pool API (GET, POST, PATCH), for example: GET /api/clusters_mgmt/v1/clusters/{cluster_id}/node_pools

• Max surge: items > management_upgrade > max_surge
• Max unavailable: items > management_upgrade > max_unavailable
• Node drain timeout: items > node_drain_grace_period > value (NOTE: this will always be in minutes)