PlanResourceChange undesirable updates

[VenelinMartinov]

What happened?

A bunch of providers got unexpeced diffs with PlanResourceChange.

Example

https://github.com/pulumi/pulumi-azuread/actions/runs/9065108544/job/24905710331?pr=1050
https://github.com/pulumi/pulumi-auth0/actions/runs/9065083805/job/24904827772?pr=540
https://github.com/pulumi/pulumi-hcloud/actions/runs/9065349562/job/24906565191?pr=506
https://github.com/pulumi/pulumi-cloudflare/actions/runs/9065158604/job/24905830152?pr=774
https://github.com/pulumi/pulumi-aws/actions/runs/9065699572/job/24909556049?pr=3931

Output of pulumi about

.

Additional context

No response

Contributing

Vote on this issue by adding a 👍 reaction.
To contribute a fix for this issue, leave a comment (and link to your pull request, if you've opened one already).

[t0yv0]

Having a quick look, some of the AWS failures look serious but we need to go one by one.

[VenelinMartinov]

A lot of the updates are due to #2047

But also we have a few others:

meta -> metadata: https://github.com/pulumi/pulumi-auth0/actions/runs/9065083805/job/24904827772?pr=540

some unexpected nulls and unknowns in AWS: https://github.com/pulumi/pulumi-aws/actions/runs/9065699572/job/24909556049?pr=3931

Also some other issues with RDS instances there

[t0yv0]

Can you send me an AWS branch to debug. Thanks.

[VenelinMartinov]

I was going to ask for help here, thanks!

pulumi/pulumi-aws#3931 is a slightly older bridge master

pulumi/pulumi-aws#3983 is the one with the block normalizations PR, feel free to ignore this one but the errors look fairly similar at a glance

You can also run the AWS provider with PULUMI_ENABLE_PLAN_RESOURCE_CHANGE=1 and it'll enable PlanResourceChange for all resources.

[t0yv0]

Perfect, I want the one with everything latest so 3983 sounds good. I'll take a look.

[VenelinMartinov]

Thanks!

[VenelinMartinov]

Aside from the AWS failures the others all look like #2047

[t0yv0]

AWS is done in pulumi/pulumi-aws#4013

[VenelinMartinov]

last round I ran here:

FAILURE pulumi/pulumi-hcloud#525
FAILURE pulumi/pulumi-auth0#564
FAILURE pulumi/pulumi-aws#4048
FAILURE pulumi/pulumi-fastly#559

hcloud is pulumi/pulumi-hcloud#528, present on master
auth0 and fastly need a resource enrolled in PRC: pulumi/pulumi-fastly#556 and pulumi/pulumi-auth0#566

AWS is explained in pulumi/pulumi-aws#4013

hence this is closed by #2065 or equivalent.

[VenelinMartinov]

This was fixed with the new refresh behaviour in the CLI.