guineveresaenger
commented
1 month ago Hi @yellowhat - I'm so sorry you're encountering this.
I've reproduced this bug on my end and we are taking a look.
Closed yellowhat closed 1 month ago
guineveresaenger
commented
1 month ago Hi @yellowhat - I'm so sorry you're encountering this.
I've reproduced this bug on my end and we are taking a look.
guineveresaenger
commented
1 month ago This issue is not limited to networking.Network. It's failing on networking.Router as well:
import * as openstack from "@pulumi/openstack";
const router1 = new openstack.networking.Router("router_1", {
name: "my_router",
adminStateUp: true,
externalNetworkId: "f67f0d72-0ddf-11e4-9d95-e1f29f417e2f",
});Other modules are failing also:
const basic = new openstack.database.User("basic", {
name: "basic",
instanceId: basicOpenstackDbInstanceV1.id,
password: "password",
databases: ["testdb"],
});update: The change that is making this panic is "caused" by a newly added call to d.GetRawConfig().GetAttr(key) on the provider configuration.
It appears that Pulumi's provider configuration isn't correctly implementing the config for this call, as at that point in the call stack, d.GetRawConfig() returns the following cty.NullVal:
d.GetRawConfig()=cty.NullVal(cty.Object(map[string]cty.Type{"allow_reauth":cty.Bool, "application_credential_id":cty.String, "application_credential_name":cty.String, "application_credential_secret":cty.String, "auth_url":cty.String, "cacert_file":cty.String, "cert":cty.String, "cloud":cty.String, "default_domain":cty.String, "delayed_auth":cty.Bool, "disable_no_cache_header":cty.Bool, "domain_id":cty.String, "domain_name":cty.String, "enable_logging":cty.Bool, "endpoint_overrides":cty.Map(cty.String), "endpoint_type":cty.String, "insecure":cty.Bool, "key":cty.String, "max_retries":cty.Number, "password":cty.String, "project_domain_id":cty.String, "project_domain_name":cty.String, "region":cty.String, "swauth":cty.Bool, "system_scope":cty.Bool, "tenant_id":cty.String, "tenant_name":cty.String, "token":cty.String, "user_domain_id":cty.String, "user_domain_name":cty.String, "user_id":cty.String, "user_name":cty.String}))I have attempted to run this program with Terraform, and am not encountering this issue - this is a bug/missing functionality on our end for the provider config. I'll file an issue in the pulumi-terraform-bridge to identify and address the root cause.
A way to hotfix this provider might be to patch up this line to use the previous version's method; however, the new function getOkExists is called in many more places and may panic similarly on resources.
Unfortunately, we're currently unable to verify actual resource creation due to an issue with our openstack account. We expect this to resolve soon, and should be able to have an update.
We recommend that folks refrain from updating to this version of the provider in the meantime, and regret the trouble.
pulumi-bot
commented
1 month ago This issue has been addressed in PR #656 and shipped in release v4.1.1.
Describe what happened
Hi, with the latest release (4.1.0) I am getting the following error on creating a
openstack.networking.Network:I would say that this is creating the issue.
Downgrading to 4.0.1 is a workaround.
Sample program
Log output
No response
Affected Resource(s)
openstack.networking.NetworkOutput of
pulumi aboutAdditional 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).