fix: to_restart and to_config in get_hosted_entity_statuses

[SteBaum]

Which issue(s) this PR fixes

Fixes None

The to_restart and to_config indicators must be true only if component is active, otherwise if it belongs to a node that has been decommissioned there is no point to restart or reconfigure it and it will fail anyways since the host is not there anymore. Therefore an additional condition has been put for these indicators to be true.

Additional comments

Agreements

• I hereby declare this contribution to be licensed under the Apache License Version 2.0, January 2004.
• I hereby agree to grant TOSIT a copyright license to use my contributions.

[PaulFarault]

Seems ok, please add a comment and/or something in the docstring.

[SteBaum]

Seems ok, please add a comment and/or something in the docstring.

Has been done.

[PaulFarault]

Well, I spoke too soon. Why make the reading of the to_restart and to_config columns depend on the is_active column?

Can't we just use the filter_active filter to only select "truly" configurable/restartable components instead of altering values on read?

[SteBaum]

Well, I spoke too soon. Why make the reading of the to_restart and to_config columns depend on the is_active column?

Can't we just use the filter_active filter to only select "truly" configurable/restartable components instead of altering values on read?

Yes, it is possible to put the filter_active to True in the script of thetdp plan reconfigure command. However, the reason I put it in the query get_hosted_entity_statuses is that I thought it would always be illogical to restart or configure a component which is on a host that does not exist anymore so it was logical to condition the values of these two columns on whether the component actually exists or not.

[PaulFarault]

Ok, just use the filter_active where it's needed then!

[SteBaum]

Ok, just use the filter_active where it's needed then!

Done

[SteBaum]

is_active column has been removed