Files
Chris CoutinhoandClaude Opus 4.7 e4c552cd19 fix(infra): address PR review feedback on tf modules
Deployer role:
- Add servicediscovery actions; module always creates Cloud Map namespace
  and service so the policy must grant CreatePrivateDnsNamespace etc.
- Make Route53 + ACM permissions unconditional. The server module always
  issues an ACM cert and writes Route53 records (no CloudFront default-cert
  path exists), so gating these on route53_zone_ids was broken. Split
  Route53 into hosted-zone management (always) plus record-set mutation
  (scoped to caller-supplied zones, falls back to *).
- Remove unused cloudfront:* statement; no CloudFront resources in module.
- Replace acm:* wildcard with explicit cert-management action set.

Server module:
- qdrant_image_tag is now nullable with default null and validated against
  use_external_qdrant, so external-qdrant callers can omit it instead of
  passing a sentinel "unused" value.
- task_role_arn and efs_id outputs marked sensitive; qdrant_dns_name returns
  null when use_external_qdrant = true.
- ALB SG now has matching IPv6 egress rule (was v4-only).
- nextcloud_url validates the https:// scheme.
- random_pet.subdomain keeper includes zone_name so a zone migration that
  preserves zone_id still triggers regeneration.
- Pin required_version >= 1.9 on both modules.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 23:24:09 +02:00

379 lines
11 KiB
Terraform

terraform {
required_version = ">= 1.9"
required_providers {
aws = {
source = "hashicorp/aws"
version = "~> 6.0"
}
}
}
data "aws_caller_identity" "current" {}
data "aws_partition" "current" {}
locals {
account_id = data.aws_caller_identity.current.account_id
partition = data.aws_partition.current.partition
prefix = var.module_name_prefix
}
###
# Role + trust
resource "aws_iam_role" "this" {
name = var.role_name
path = var.role_path
assume_role_policy = data.aws_iam_policy_document.trust.json
description = "Deploy + manage the nextcloud-mcp-server Terraform module."
}
data "aws_iam_policy_document" "trust" {
statement {
sid = "AllowTrustedPrincipals"
actions = ["sts:AssumeRole"]
principals {
type = "AWS"
identifiers = var.trusted_principal_arns
}
}
}
###
# Deployer policy
#
# Scoping strategy: actions that have no usable resource-ARN form at create
# time (RegisterTaskDefinition, CreateLoadBalancer, CreateFileSystem, all of
# CloudFront, etc.) are granted on `*` — IAM gives no other option there.
# Where ARN scoping IS available and worth the surface reduction (IAM, logs,
# secrets, route53), the policy is scoped tight to the module's name prefix.
data "aws_iam_policy_document" "deployer" {
# --- Compute / runtime: ECS, ALB, EFS ---
#
# All three services are gated by SG/SCP/account boundaries already, and
# most of their create-time APIs reject resource-ARN scoping. Granting
# service-wide is the standard pattern for a deployer role.
statement {
sid = "EcsService"
actions = ["ecs:*"]
resources = ["*"]
}
statement {
sid = "ElbV2Service"
actions = ["elasticloadbalancing:*"]
resources = ["*"]
}
statement {
sid = "EfsService"
actions = ["elasticfilesystem:*"]
resources = ["*"]
}
# --- Service Discovery / Cloud Map ---
#
# Module always creates `aws_service_discovery_private_dns_namespace` (in
# service_discovery.tf) and `aws_service_discovery_service` (in qdrant.tf).
# Cloud Map APIs don't accept resource-ARN scoping at create time; the
# blast radius is bounded by the SG/account boundary already.
statement {
sid = "ServiceDiscoveryService"
actions = [
"servicediscovery:CreatePrivateDnsNamespace",
"servicediscovery:DeleteNamespace",
"servicediscovery:GetNamespace",
"servicediscovery:ListNamespaces",
"servicediscovery:GetOperation",
"servicediscovery:CreateService",
"servicediscovery:DeleteService",
"servicediscovery:GetService",
"servicediscovery:UpdateService",
"servicediscovery:ListServices",
"servicediscovery:TagResource",
"servicediscovery:UntagResource",
"servicediscovery:ListTagsForResource",
]
resources = ["*"]
}
# --- Logs ---
#
# Module creates a single log group: `/ecs/${var.name}`. Scope CW Logs
# writes/reads to that prefix; describe APIs need `*` because they don't
# accept resource-ARN scoping.
statement {
sid = "LogsManageGroup"
actions = ["logs:*"]
resources = [
"arn:${local.partition}:logs:*:${local.account_id}:log-group:/ecs/${local.prefix}*",
"arn:${local.partition}:logs:*:${local.account_id}:log-group:/ecs/${local.prefix}*:*",
"arn:${local.partition}:logs:*:${local.account_id}:log-group:/ecs/${local.prefix}*:log-stream:*",
]
}
statement {
sid = "LogsDescribe"
actions = [
"logs:DescribeLogGroups",
"logs:DescribeLogStreams",
]
resources = ["*"]
}
# --- IAM ---
#
# The module creates two roles under path `/ecs/`: `${name}-execution` and
# `${name}-task`, plus inline policies on each. Scope to that path+prefix
# so the deployer can't pivot to creating arbitrary roles.
statement {
sid = "IamManageModuleRoles"
actions = [
"iam:CreateRole",
"iam:GetRole",
"iam:DeleteRole",
"iam:UpdateRole",
"iam:UpdateAssumeRolePolicy",
"iam:PutRolePolicy",
"iam:GetRolePolicy",
"iam:DeleteRolePolicy",
"iam:ListRolePolicies",
"iam:ListAttachedRolePolicies",
"iam:TagRole",
"iam:UntagRole",
"iam:ListRoleTags",
]
resources = [
"arn:${local.partition}:iam::${local.account_id}:role/ecs/${local.prefix}-*",
]
}
# ECS RunTask + service updates need iam:PassRole on the task/exec roles.
statement {
sid = "IamPassModuleRoles"
actions = ["iam:PassRole"]
resources = [
"arn:${local.partition}:iam::${local.account_id}:role/ecs/${local.prefix}-*",
]
condition {
test = "StringEquals"
variable = "iam:PassedToService"
values = ["ecs-tasks.amazonaws.com"]
}
}
# The execution role attaches the AWS-managed AmazonECSTaskExecutionRolePolicy.
# Restrict the Attach/Detach actions to that single managed policy ARN so
# this grant can't be used to attach AdministratorAccess or similar.
statement {
sid = "IamAttachManagedTaskExecPolicy"
actions = [
"iam:AttachRolePolicy",
"iam:DetachRolePolicy",
]
resources = [
"arn:${local.partition}:iam::${local.account_id}:role/ecs/${local.prefix}-*",
]
condition {
test = "ArnEquals"
variable = "iam:PolicyARN"
values = [
"arn:${local.partition}:iam::aws:policy/service-role/AmazonECSTaskExecutionRolePolicy",
]
}
}
# --- Secrets Manager ---
#
# The module reads a caller-supplied secret ARN at deploy time
# (execution-role policy in iam.tf:26). Optionally allow create/update
# for callers who manage the secret in their own TF.
statement {
sid = "SecretsRead"
actions = [
"secretsmanager:GetSecretValue",
"secretsmanager:DescribeSecret",
"secretsmanager:ListSecrets",
]
resources = [
"arn:${local.partition}:secretsmanager:*:${local.account_id}:secret:${var.secret_name_prefix}*",
]
}
dynamic "statement" {
for_each = var.allow_secret_create ? [1] : []
content {
sid = "SecretsManage"
actions = [
"secretsmanager:CreateSecret",
"secretsmanager:UpdateSecret",
"secretsmanager:DeleteSecret",
"secretsmanager:PutSecretValue",
"secretsmanager:TagResource",
"secretsmanager:UntagResource",
"secretsmanager:GetResourcePolicy",
"secretsmanager:PutResourcePolicy",
]
resources = [
"arn:${local.partition}:secretsmanager:*:${local.account_id}:secret:${var.secret_name_prefix}*",
]
}
}
# --- EC2: Security Groups + describe-only networking ---
#
# SG mutations have no usable resource-ARN scoping at create time
# (CreateSecurityGroup returns the ID), so SG actions are granted on `*`.
# The describe set is needed for the module's data sources and SG-rule
# references (managed prefix list lookup for the CloudFront SG lock).
statement {
sid = "Ec2NetworkDescribe"
actions = [
"ec2:DescribeVpcs",
"ec2:DescribeSubnets",
"ec2:DescribeAvailabilityZones",
"ec2:DescribeSecurityGroups",
"ec2:DescribeSecurityGroupRules",
"ec2:DescribeNetworkInterfaces",
"ec2:DescribeRouteTables",
"ec2:DescribePrefixLists",
"ec2:DescribeManagedPrefixLists",
"ec2:GetManagedPrefixListEntries",
"ec2:DescribeTags",
]
resources = ["*"]
}
statement {
sid = "Ec2SecurityGroupManage"
actions = [
"ec2:CreateSecurityGroup",
"ec2:DeleteSecurityGroup",
"ec2:ModifySecurityGroupRules",
"ec2:AuthorizeSecurityGroupIngress",
"ec2:AuthorizeSecurityGroupEgress",
"ec2:RevokeSecurityGroupIngress",
"ec2:RevokeSecurityGroupEgress",
"ec2:UpdateSecurityGroupRuleDescriptionsIngress",
"ec2:UpdateSecurityGroupRuleDescriptionsEgress",
"ec2:CreateTags",
"ec2:DeleteTags",
]
resources = ["*"]
}
# --- Route53 ---
#
# The server module always creates ACM cert + Route53 records (alias to ALB
# and DNS-01 validation), so these statements are unconditional. Cloud Map's
# CreatePrivateDnsNamespace also creates a Route53 private hosted zone
# under the caller's identity, which needs the hosted-zone management
# actions below.
# Hosted-zone CRUD (private zones for Cloud Map, public-zone reads for the
# caller-supplied zone). ChangeResourceRecordSets is the destructive action
# and is split into a separate statement scoped to caller-supplied zones.
statement {
sid = "Route53HostedZoneManage"
actions = [
"route53:CreateHostedZone",
"route53:GetHostedZone",
"route53:DeleteHostedZone",
"route53:ListHostedZones",
"route53:ListHostedZonesByVPC",
"route53:AssociateVPCWithHostedZone",
"route53:DisassociateVPCFromHostedZone",
"route53:ChangeTagsForResource",
"route53:ListTagsForResource",
"route53:ListResourceRecordSets",
]
resources = ["*"]
}
# ChangeResourceRecordSets is scoped to caller-supplied public zones when
# `route53_zone_ids` is set. Falls back to `*` only as a convenience for
# callers who haven't enumerated their zones — strongly recommend setting
# the variable.
statement {
sid = "Route53RecordsForZones"
actions = ["route53:ChangeResourceRecordSets"]
resources = (
length(var.route53_zone_ids) > 0
? [for zone_id in var.route53_zone_ids : "arn:${local.partition}:route53:::hostedzone/${zone_id}"]
: ["*"]
)
}
# GetChange takes a change-id, not a zone ARN — must be `*`.
statement {
sid = "Route53GetChange"
actions = ["route53:GetChange"]
resources = ["*"]
}
# --- ACM ---
#
# The server module always issues an ACM cert. ACM cert ARNs are only
# known after RequestCertificate, so the destructive actions can't be
# scoped further at policy-write time. Restrict the action set instead of
# granting `acm:*`.
statement {
sid = "AcmManageCertificates"
actions = [
"acm:RequestCertificate",
"acm:DescribeCertificate",
"acm:DeleteCertificate",
"acm:ListCertificates",
"acm:ListTagsForCertificate",
"acm:AddTagsToCertificate",
"acm:RemoveTagsFromCertificate",
"acm:UpdateCertificateOptions",
]
resources = ["*"]
}
# --- KMS describe (AWS-managed keys for default EFS / Secrets encryption) ---
statement {
sid = "KmsDescribeAwsManaged"
actions = [
"kms:DescribeKey",
"kms:ListAliases",
]
resources = ["*"]
}
# --- Bedrock (model discovery) ---
#
# The task role grants `bedrock:InvokeModel` at runtime. The deployer
# itself doesn't invoke; it only needs to validate the model exists when
# rendering the task-role policy document. Read-only.
statement {
sid = "BedrockDescribe"
actions = [
"bedrock:GetFoundationModel",
"bedrock:ListFoundationModels",
]
resources = ["*"]
}
# --- STS ---
statement {
sid = "StsCallerIdentity"
actions = ["sts:GetCallerIdentity"]
resources = ["*"]
}
}
resource "aws_iam_policy" "deployer" {
name = "${var.role_name}-policy"
path = var.role_path
description = "Least-privilege policy for the nextcloud-mcp-server deployer role."
policy = data.aws_iam_policy_document.deployer.json
}
resource "aws_iam_role_policy_attachment" "deployer" {
role = aws_iam_role.this.name
policy_arn = aws_iam_policy.deployer.arn
}