Error: Failed to update proposed state from prior state #2487
Labels
Comments
|
Having exactly the same issue today. No change of anything except provider version upgraded to |
|
Same here, looks related to f83d63a Trace is : for this |
|
Temporary workaround is using the previous provider version. (verified working) |
|
Hi, would like to add on to this issue since this is referencing Kibana! I too faced the issue, kubernetes_manifest.my_kibana_dev: Modifying...
╷
│ Error: Provider produced inconsistent result after apply
│
│ When applying changes to kubernetes_manifest.my_kibana_dev, provider
│ "provider[\"registry.terraform.io/hashicorp/kubernetes\"]" produced an
│ unexpected new value: .object: wrong final value type: incorrect object
│ attributes.
│
│ This is a bug in the provider, which should be reported in the provider's
│ own issue tracker.
╵
Operation failed: failed running terraform apply (exit 1)but with a different setup (using ECK operator) TLDR;
|
|
I am seeing the same error when trying to update FluxCD
The error thrown is: EDIT: happens with v2.31 as well as with v2.29.0 |
|
Hi guys! Ok I managed to fix this upon looking at the kubernetes docs more intensely. Pay attention to this part of the docs regarding computed fields TLDR;
UPDATE AttributeName("spec"): can't use tftypes.Object["config":tftypes.DynamicPseudoType, "count":tftypes.Number, "elasticsearchRef":tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String], "enterpriseSearchRef":tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String], "http":tftypes.Object["service":tftypes.Object["metadata":tftypes.Object["annotations":tftypes.Map[tftypes.String], "finalizers":tftypes.List[tftypes.String], "labels":tftypes.Map[tftypes.String], "name":tftypes.String, "namespace":tftypes.String], "spec":tftypes.Object["allocateLoadBalancerNodePorts":tftypes.Bool, "clusterIP":tftypes.String, "clusterIPs":tftypes.List[tftypes.String], "externalIPs":tftypes.List[tftypes.String], "externalName":tftypes.String, "externalTrafficPolicy":tftypes.String, "healthCheckNodePort":tftypes.Number, "internalTrafficPolicy":tftypes.String, "ipFamilies":tftypes.List[tftypes.String], "ipFamilyPolicy":tftypes.String, "loadBalancerClass":tftypes.String, "loadBalancerIP":tftypes.String, "loadBalancerSourceRanges":tftypes.List[tftypes.String], "ports":tftypes.List[tftypes.Object["appProtocol":tftypes.String, "name":tftypes.String, "nodePort":tftypes.Number, "port":tftypes.Number, "protocol":tftypes.String, "targetPort":tftypes.String]], "publishNotReadyAddresses":tftypes.Bool, "selector":tftypes.Map[tftypes.String], "sessionAffinity":tftypes.String, "sessionAffinityConfig":tftypes.Object["clientIP":tftypes.Object["timeoutSeconds":tftypes.Number]], "type":tftypes.String]], "tls":tftypes.Object["certificate":tftypes.Object["secretName":tftypes.String], "selfSignedCertificate":tftypes.Object["disabled":tftypes.Bool, "subjectAltNames":tftypes.List[tftypes.Object["dns":tftypes.String, "ip":tftypes.String]]]]], "image":tftypes.String, "monitoring":tftypes.Object["logs":tftypes.Object["elasticsearchRefs":tftypes.List[tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String]]], "metrics":tftypes.Object["elasticsearchRefs":tftypes.List[tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String]]]], "podTemplate":tftypes.Object["metadata":tftypes.Object["creationTimestamp":tftypes.DynamicPseudoType], "spec":tftypes.Object["automountServiceAccountToken":tftypes.Bool, "containers":tftypes.Tuple[tftypes.Object["name":tftypes.String, "resources":tftypes.Object["limits":tftypes.Object["cpu":tftypes.String, "memory":tftypes.String]]]], "nodeSelector":tftypes.Object["nodepool":tftypes.String]]], "revisionHistoryLimit":tftypes.Number, "secureSettings":tftypes.List[tftypes.Object["entries":tftypes.List[tftypes.Object["key":tftypes.String, "path":tftypes.String]], "secretName":tftypes.String]], "serviceAccountName":tftypes.String, "version":tftypes.String] as tftypes.Object["config":tftypes.DynamicPseudoType, "count":tftypes.Number, "elasticsearchRef":tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String], "enterpriseSearchRef":tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String], "http":tftypes.Object["service":tftypes.Object["metadata":tftypes.Object["annotations":tftypes.Map[tftypes.String], "finalizers":tftypes.List[tftypes.String], "labels":tftypes.Map[tftypes.String], "name":tftypes.String, "namespace":tftypes.String], "spec":tftypes.Object["allocateLoadBalancerNodePorts":tftypes.Bool, "clusterIP":tftypes.String, "clusterIPs":tftypes.List[tftypes.String], "externalIPs":tftypes.List[tftypes.String], "externalName":tftypes.String, "externalTrafficPolicy":tftypes.String, "healthCheckNodePort":tftypes.Number, "internalTrafficPolicy":tftypes.String, "ipFamilies":tftypes.List[tftypes.String], "ipFamilyPolicy":tftypes.String, "loadBalancerClass":tftypes.String, "loadBalancerIP":tftypes.String, "loadBalancerSourceRanges":tftypes.List[tftypes.String], "ports":tftypes.List[tftypes.Object["appProtocol":tftypes.String, "name":tftypes.String, "nodePort":tftypes.Number, "port":tftypes.Number, "protocol":tftypes.String, "targetPort":tftypes.String]], "publishNotReadyAddresses":tftypes.Bool, "selector":tftypes.Map[tftypes.String], "sessionAffinity":tftypes.String, "sessionAffinityConfig":tftypes.Object["clientIP":tftypes.Object["timeoutSeconds":tftypes.Number]], "type":tftypes.String]], "tls":tftypes.Object["certificate":tftypes.Object["secretName":tftypes.String], "selfSignedCertificate":tftypes.Object["disabled":tftypes.Bool, "subjectAltNames":tftypes.List[tftypes.Object["dns":tftypes.String, "ip":tftypes.String]]]]], "image":tftypes.String, "monitoring":tftypes.Object["logs":tftypes.Object["elasticsearchRefs":tftypes.List[tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String]]], "metrics":tftypes.Object["elasticsearchRefs":tftypes.List[tftypes.Object["name":tftypes.String, "namespace":tftypes.String, "secretName":tftypes.String, "serviceName":tftypes.String]]]], "podTemplate":tftypes.Object["spec":tftypes.Object["automountServiceAccountToken":tftypes.Bool, "containers":tftypes.Tuple[tftypes.Object["name":tftypes.String, "resources":tftypes.Object["limits":tftypes.Object["cpu":tftypes.String, "memory":tftypes.String]]]], "nodeSelector":tftypes.Object["nodepool":tftypes.String]]], "revisionHistoryLimit":tftypes.Number, "secureSettings":tftypes.List[tftypes.Object["entries":tftypes.List[tftypes.Object["key":tftypes.String, "path":tftypes.String]], "secretName":tftypes.String]], "serviceAccountName":tftypes.String, "version":tftypes.String]ProcessUpon looking at the docs, I first figured out that it would probably be a great idea to look into the resulting apiVersion: redacted
kind: Kibana
metadata:
annotations:
association.k8s.elastic.co/es-conf: '{redacted}'
creationTimestamp: redacted
generation: redacted
name: redacted
namespace: redacted
resourceVersion: redacted
uid: redacted
spec:
count: redacted
elasticsearchRef:
name: redacted
enterpriseSearchRef: {redacted}
http:
service:
metadata: {redacted}
spec: {redacted}
tls:
certificate: {redacted}
monitoring:
logs: {redacted}
metrics: {redacted}
podTemplate:
metadata:
creationTimestamp: redacted
spec:
automountServiceAccountToken: redacted
containers:
- name: kibana
resources:
limits:
cpu: redacted
memory: redacted
nodeSelector:
nodepool: redacted
version: redacted
status:
associationStatus: redacted
count: redacted
elasticsearchAssociationStatus: redacted
health: redacted
observedGeneration: redacted
selector: redacted
version: redactedI found the above really weird because some fields were missing in my original resource "kubernetes_manifest" "redacted" {
manifest = {
apiVersion = "redacted"
kind = "Kibana"
metadata = {
name = redacted
namespace = redacted
}
spec = {
version = redacted
count = redacted
elasticsearchRef = {
name = redacted
}
podTemplate = {
spec = {
automountServiceAccountToken = redacted
containers = [
{
name = "kibana"
resources = {
limits = {
memory = redacted
cpu = redacted
}
}
}
]
nodeSelector = {
nodepool = redacted
}
}
}
}
}
field_manager {
...
}
depends_on = [redacted]
}I thought perhaps the fields that ended up being changed were not in my resource "kubernetes_manifest" "redacted" {
manifest = {
...
}
# Added this block
computed_fields = [
"metadata.annotations",
"metadata.creationTimestamp",
"metadata.generation",
"metadata.resourceVersion",
"metadata.uid",
"spec.enterpriseSearchRef",
"spec.http",
"spec.monitoring",
"spec.podTemplate.metadata",
"status"
]
field_manager {
...
}
depends_on = [redacted]
}II thought that by adding the fields I did not mutate in my kubernetes_manifest.lome_kibana_sgp1_dev: Modifying...
╷
│ Error: Provider produced inconsistent result after apply
│
│ When applying changes to kubernetes_manifest.lome_kibana_sgp1_dev, provider
│ "provider[\"registry.terraform.io/hashicorp/kubernetes\"]" produced an
│ unexpected new value: .object: wrong final value type: incorrect object
│ attributes.
│
│ This is a bug in the provider, which should be reported in the provider's
│ own issue tracker.
╵
Operation failed: failed running terraform apply (exit 1)Which got me thinking that perhaps, there were some fields in the resource "kubernetes_manifest" "redacted" {
manifest = {
...
}
# Changed this block
computed_fields = [
"metadata",
"spec",
"status"
]
field_manager {
...
}
depends_on = [redacted]
}Which did the trick for me. Final Thoughts
|
|
I have the same issue as described above, neither field_manager to terraform, nor putting everything in computed_fields works for me. Version 2.29.0 has the same issue. I always endup with : My terraform file : I only have the issue with wait section, but I think this is because it first think it is deployed and resolves before encountering the error. Here are the full trace logs : https://gist.github.com/685f48c73cd814eb7fe5375fe85ca48e.git |
|
Just encountered this issue on v2.34.0. |
|
Just encountered this issue on v2.35.0 |
|
Same issue encountered on v2.35.1 with elasticsearch + kibana. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I were using 2.29.0 version and after updating it to 2.30.0 version I got this kind of error, without changing anything, only provider version
Error: Failed to update proposed state from prior state
with kubernetes_manifest.kibana,
on main.tf line 87, in resource "kubernetes_manifest" "kibana":
87: resource "kubernetes_manifest" "kibana" {
AttributeName("podTemplate"): can't use
tftypes.Object["metadata":tftypes.Object["creationTimestamp":tftypes.DynamicPseudoType],
"spec":tftypes.Object["containers":tftypes.Tuple[tftypes.Object["env":tftypes.Tuple[tftypes.Object["name":tftypes.String,
"value":tftypes.String]], "name":tftypes.String,
"readinessProbe":tftypes.Object["httpGet":tftypes.Object["path":tftypes.String,
"port":tftypes.Number, "scheme":tftypes.String]],
The text was updated successfully, but these errors were encountered: