-
Notifications
You must be signed in to change notification settings - Fork 1
feat: cluster credential rotator #19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Draft
tuunit
wants to merge
1
commit into
main
Choose a base branch
from
feat/cluster-credential-rotator
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Draft
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
386 changes: 386 additions & 0 deletions
386
bootstrap/platform-components/helm/argo-cd/script/cluster-credential-rotator.tmpl
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,386 @@ | ||
| #!/bin/sh | ||
|
|
||
| set -u | ||
|
|
||
| RELEASE_NAMESPACE="{{ .Release.Namespace }}" | ||
| LABEL_KEY="{{ .Values.clusterCredentialRotator.secret.labelKey }}" | ||
| LABEL_VALUE="{{ .Values.clusterCredentialRotator.secret.labelValue }}" | ||
| REQUIRED_TYPE="{{ .Values.clusterCredentialRotator.secret.requiredType }}" | ||
| BOOTSTRAP_KUBECONFIG_KEY="{{ .Values.clusterCredentialRotator.secret.bootstrapKubeconfigKey }}" | ||
| GENERATED_CONFIG_KEY="{{ .Values.clusterCredentialRotator.secret.generatedConfigKey }}" | ||
| REMOTE_NAMESPACE="{{ .Values.clusterCredentialRotator.remoteAccess.namespace }}" | ||
| REMOTE_SERVICE_ACCOUNT_NAME="{{ .Values.clusterCredentialRotator.remoteAccess.serviceAccountName }}" | ||
| REMOTE_CLUSTER_ROLE_BINDING_NAME="{{ .Values.clusterCredentialRotator.remoteAccess.clusterRoleBindingName }}" | ||
| REMOTE_CLUSTER_ROLE_NAME="{{ .Values.clusterCredentialRotator.remoteAccess.clusterRoleName }}" | ||
| TOKEN_DURATION="{{ .Values.clusterCredentialRotator.token.duration }}" | ||
| REFRESH_BEFORE="{{ .Values.clusterCredentialRotator.token.refreshBefore }}" | ||
|
|
||
| ISSUED_AT_ANNOTATION="kubara.io/cluster-credential-issued-at" | ||
| EXPIRES_AT_ANNOTATION="kubara.io/cluster-credential-expires-at" | ||
| MANAGED_BY_ANNOTATION="kubara.io/cluster-credential-managed-by" | ||
| MANAGED_BY_VALUE="cluster-credential-rotator" | ||
|
|
||
| log() { | ||
| printf '%s %s\n' "$(date -u +%Y-%m-%dT%H:%M:%SZ)" "$*" | ||
| } | ||
|
|
||
| normalize_optional_value() { | ||
| if [ "${1:-}" = "<no value>" ] || [ "${1:-}" = "<nil>" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| printf '%s' "${1:-}" | ||
| } | ||
|
|
||
| duration_to_seconds() { | ||
| awk -v value="$1" ' | ||
| function fail() { exit 1 } | ||
| BEGIN { | ||
| rest = value | ||
| total = 0 | ||
|
|
||
| if (rest == "") { | ||
| fail() | ||
| } | ||
|
|
||
| while (rest != "") { | ||
| if (match(rest, /^[0-9]+[smhd]/) == 0) { | ||
| fail() | ||
| } | ||
|
|
||
| part = substr(rest, RSTART, RLENGTH) | ||
| amount = substr(part, 1, length(part) - 1) + 0 | ||
| unit = substr(part, length(part), 1) | ||
|
|
||
| if (unit == "s") { | ||
| multiplier = 1 | ||
| } else if (unit == "m") { | ||
| multiplier = 60 | ||
| } else if (unit == "h") { | ||
| multiplier = 3600 | ||
| } else if (unit == "d") { | ||
| multiplier = 86400 | ||
| } else { | ||
| fail() | ||
| } | ||
|
|
||
| total += amount * multiplier | ||
| rest = substr(rest, RLENGTH + 1) | ||
| } | ||
|
|
||
| print total | ||
| } | ||
| ' </dev/null | ||
| } | ||
|
|
||
| json_escape() { | ||
| printf '%s' "$1" | awk ' | ||
| BEGIN { ORS = "" } | ||
| { | ||
| gsub(/\\/, "\\\\") | ||
| gsub(/"/, "\\\"") | ||
| gsub(/\r/, "\\r") | ||
| gsub(/\t/, "\\t") | ||
| printf "%s", $0 | ||
| } | ||
| ' | ||
| } | ||
|
|
||
| get_secret_annotation() { | ||
| template_string="$(printf '%s%s index .metadata.annotations "%s" %s%s' '{' '{' "$2" '}' '}')" | ||
| normalize_optional_value "$(kubectl -n "$RELEASE_NAMESPACE" get secret "$1" -o "go-template=${template_string}" 2>/dev/null || true)" | ||
| } | ||
|
|
||
| get_secret_data() { | ||
| template_string="$(printf '%s%s index .data "%s" %s%s' '{' '{' "$2" '}' '}')" | ||
| normalize_optional_value "$(kubectl -n "$RELEASE_NAMESPACE" get secret "$1" -o "go-template=${template_string}" 2>/dev/null || true)" | ||
| } | ||
|
|
||
| get_secret_value() { | ||
| encoded_value="$(get_secret_data "$1" "$2")" | ||
|
|
||
| if [ -z "$encoded_value" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| printf '%s' "$encoded_value" | base64 -d | ||
| } | ||
|
|
||
| get_secret_type() { | ||
| template_string="$(printf '%s%s .type %s%s' '{' '{' '}' '}')" | ||
| normalize_optional_value "$(kubectl -n "$RELEASE_NAMESPACE" get secret "$1" -o "go-template=${template_string}" 2>/dev/null || true)" | ||
| } | ||
|
|
||
| build_generated_kubeconfig() { | ||
| kubeconfig_path=$1 | ||
| server=$2 | ||
| ca_data=$3 | ||
| insecure=$4 | ||
| token=$5 | ||
|
|
||
| if [ "$insecure" = "true" ]; then | ||
| cat >"$kubeconfig_path" <<EOF | ||
| apiVersion: v1 | ||
| kind: Config | ||
| clusters: | ||
| - cluster: | ||
| insecure-skip-tls-verify: true | ||
| server: ${server} | ||
| name: managed-cluster | ||
| contexts: | ||
| - context: | ||
| cluster: managed-cluster | ||
| user: managed-service-account | ||
| name: managed-cluster | ||
| current-context: managed-cluster | ||
| users: | ||
| - name: managed-service-account | ||
| user: | ||
| token: ${token} | ||
| EOF | ||
| return 0 | ||
| fi | ||
|
|
||
| cat >"$kubeconfig_path" <<EOF | ||
| apiVersion: v1 | ||
| kind: Config | ||
| clusters: | ||
| - cluster: | ||
| certificate-authority-data: ${ca_data} | ||
| server: ${server} | ||
| name: managed-cluster | ||
| contexts: | ||
| - context: | ||
| cluster: managed-cluster | ||
| user: managed-service-account | ||
| name: managed-cluster | ||
| current-context: managed-cluster | ||
| users: | ||
| - name: managed-service-account | ||
| user: | ||
| token: ${token} | ||
| EOF | ||
| } | ||
|
|
||
| generated_config_token() { | ||
| printf '%s' "$1" | sed -n 's/.*"bearerToken":"\([^"]*\)".*/\1/p' | ||
| } | ||
|
|
||
| validate_generated_credentials() { | ||
| validation_kubeconfig=$1 | ||
| server=$2 | ||
| ca_data=$3 | ||
| insecure=$4 | ||
| token=$5 | ||
|
|
||
| build_generated_kubeconfig "$validation_kubeconfig" "$server" "$ca_data" "$insecure" "$token" | ||
| kubectl --kubeconfig "$validation_kubeconfig" get namespace kube-system >/dev/null 2>&1 | ||
| } | ||
|
|
||
| ensure_remote_access() { | ||
| bootstrap_kubeconfig=$1 | ||
|
|
||
| cat <<EOF | kubectl --kubeconfig "$bootstrap_kubeconfig" apply -f - >/dev/null | ||
| apiVersion: v1 | ||
| kind: Namespace | ||
| metadata: | ||
| name: ${REMOTE_NAMESPACE} | ||
| --- | ||
| apiVersion: v1 | ||
| kind: ServiceAccount | ||
| metadata: | ||
| name: ${REMOTE_SERVICE_ACCOUNT_NAME} | ||
| namespace: ${REMOTE_NAMESPACE} | ||
| --- | ||
| apiVersion: rbac.authorization.k8s.io/v1 | ||
| kind: ClusterRoleBinding | ||
| metadata: | ||
| name: ${REMOTE_CLUSTER_ROLE_BINDING_NAME} | ||
| subjects: | ||
| - kind: ServiceAccount | ||
| name: ${REMOTE_SERVICE_ACCOUNT_NAME} | ||
| namespace: ${REMOTE_NAMESPACE} | ||
| roleRef: | ||
| apiGroup: rbac.authorization.k8s.io | ||
| kind: ClusterRole | ||
| name: ${REMOTE_CLUSTER_ROLE_NAME} | ||
| EOF | ||
| } | ||
|
|
||
| secret_needs_refresh() { | ||
| secret_name=$1 | ||
| validation_kubeconfig=$2 | ||
| server=$3 | ||
| ca_data=$4 | ||
| insecure=$5 | ||
| refresh_after_epoch=$6 | ||
|
|
||
| current_config="$(get_secret_value "$secret_name" "$GENERATED_CONFIG_KEY")" | ||
| if [ -z "$current_config" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| expires_at="$(get_secret_annotation "$secret_name" "$EXPIRES_AT_ANNOTATION")" | ||
| if [ -z "$expires_at" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| case "$expires_at" in | ||
| ''|*[!0-9]*) | ||
| return 0 | ||
| ;; | ||
| esac | ||
|
|
||
| if [ "$refresh_after_epoch" -ge "$expires_at" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| current_token="$(generated_config_token "$current_config")" | ||
| if [ -z "$current_token" ]; then | ||
| return 0 | ||
| fi | ||
|
|
||
| if ! validate_generated_credentials "$validation_kubeconfig" "$server" "$ca_data" "$insecure" "$current_token"; then | ||
| return 0 | ||
| fi | ||
|
|
||
| return 1 | ||
| } | ||
|
|
||
| patch_secret_with_credentials() { | ||
| secret_name=$1 | ||
| patch_file=$2 | ||
| cluster_name=$3 | ||
| server=$4 | ||
| generated_config=$5 | ||
| issued_at=$6 | ||
| expires_at=$7 | ||
|
|
||
| escaped_name="$(json_escape "$cluster_name")" | ||
| escaped_server="$(json_escape "$server")" | ||
| escaped_config="$(json_escape "$generated_config")" | ||
|
|
||
| cat >"$patch_file" <<EOF | ||
| { | ||
| "metadata": { | ||
| "labels": { | ||
| "argocd.argoproj.io/secret-type": "cluster", | ||
| "${LABEL_KEY}": "${LABEL_VALUE}" | ||
| }, | ||
| "annotations": { | ||
| "${ISSUED_AT_ANNOTATION}": "${issued_at}", | ||
| "${EXPIRES_AT_ANNOTATION}": "${expires_at}", | ||
| "${MANAGED_BY_ANNOTATION}": "${MANAGED_BY_VALUE}" | ||
| } | ||
| }, | ||
| "stringData": { | ||
| "name": "${escaped_name}", | ||
| "server": "${escaped_server}", | ||
| "${GENERATED_CONFIG_KEY}": "${escaped_config}" | ||
| } | ||
| } | ||
| EOF | ||
|
|
||
| kubectl -n "$RELEASE_NAMESPACE" patch secret "$secret_name" --type merge --patch-file "$patch_file" >/dev/null | ||
| } | ||
|
|
||
| reconcile_secret() ( | ||
| set -u | ||
|
|
||
| secret_name=$1 | ||
| bootstrap_kubeconfig_file=$(mktemp) | ||
| validation_kubeconfig_file=$(mktemp) | ||
| patch_file=$(mktemp) | ||
| trap 'rm -f "$bootstrap_kubeconfig_file" "$validation_kubeconfig_file" "$patch_file"' EXIT INT TERM | ||
|
|
||
| secret_type="$(get_secret_type "$secret_name")" | ||
| if [ -n "$REQUIRED_TYPE" ] && [ "$secret_type" != "$REQUIRED_TYPE" ]; then | ||
| log "skipping ${secret_name}: expected secret type ${REQUIRED_TYPE}, found ${secret_type:-<empty>}" | ||
| return 0 | ||
| fi | ||
|
|
||
| bootstrap_kubeconfig="$(get_secret_value "$secret_name" "$BOOTSTRAP_KUBECONFIG_KEY")" | ||
| if [ -z "$bootstrap_kubeconfig" ]; then | ||
| log "secret ${secret_name} is missing required key ${BOOTSTRAP_KUBECONFIG_KEY}" | ||
| return 1 | ||
| fi | ||
|
|
||
| printf '%s' "$bootstrap_kubeconfig" >"$bootstrap_kubeconfig_file" | ||
|
|
||
| cluster_name="$(get_secret_value "$secret_name" "name")" | ||
| if [ -z "$cluster_name" ]; then | ||
| cluster_name="$secret_name" | ||
| fi | ||
|
|
||
| server="$(get_secret_value "$secret_name" "server")" | ||
| if [ -z "$server" ]; then | ||
| server="$(kubectl --kubeconfig "$bootstrap_kubeconfig_file" config view --raw --minify -o jsonpath='{.clusters[0].cluster.server}')" | ||
| fi | ||
|
|
||
| if [ -z "$server" ]; then | ||
| log "secret ${secret_name} did not yield a cluster server endpoint" | ||
| return 1 | ||
| fi | ||
|
|
||
| ca_data="$(kubectl --kubeconfig "$bootstrap_kubeconfig_file" config view --raw --minify -o jsonpath='{.clusters[0].cluster.certificate-authority-data}')" | ||
| insecure="$(kubectl --kubeconfig "$bootstrap_kubeconfig_file" config view --raw --minify -o jsonpath='{.clusters[0].cluster.insecure-skip-tls-verify}')" | ||
| if [ -z "$insecure" ]; then | ||
| insecure="false" | ||
| fi | ||
|
|
||
| current_epoch="$(date +%s)" | ||
| refresh_after_epoch=$((current_epoch + REFRESH_BEFORE_SECONDS)) | ||
|
|
||
| if ! secret_needs_refresh "$secret_name" "$validation_kubeconfig_file" "$server" "$ca_data" "$insecure" "$refresh_after_epoch"; then | ||
| log "secret ${secret_name} already has valid generated credentials" | ||
| return 0 | ||
| fi | ||
|
|
||
| log "refreshing credentials for ${secret_name}" | ||
|
|
||
| ensure_remote_access "$bootstrap_kubeconfig_file" | ||
| token="$(kubectl --kubeconfig "$bootstrap_kubeconfig_file" -n "$REMOTE_NAMESPACE" create token "$REMOTE_SERVICE_ACCOUNT_NAME" --duration "$TOKEN_DURATION")" | ||
|
|
||
| if [ -z "$token" ]; then | ||
| log "failed to mint a service account token for ${secret_name}" | ||
| return 1 | ||
| fi | ||
|
|
||
| generated_config="$(printf '{"bearerToken":"%s","tlsClientConfig":{"caData":"%s","insecure":%s}}' "$token" "$ca_data" "$insecure")" | ||
| issued_at="$current_epoch" | ||
| expires_at=$((current_epoch + TOKEN_DURATION_SECONDS)) | ||
|
|
||
| patch_secret_with_credentials "$secret_name" "$patch_file" "$cluster_name" "$server" "$generated_config" "$issued_at" "$expires_at" | ||
|
|
||
| if ! validate_generated_credentials "$validation_kubeconfig_file" "$server" "$ca_data" "$insecure" "$token"; then | ||
| log "generated credentials for ${secret_name} failed validation after refresh" | ||
| return 1 | ||
| fi | ||
|
|
||
| log "successfully refreshed credentials for ${secret_name}" | ||
| ) | ||
|
|
||
| TOKEN_DURATION_SECONDS="$(duration_to_seconds "$TOKEN_DURATION")" | ||
| REFRESH_BEFORE_SECONDS="$(duration_to_seconds "$REFRESH_BEFORE")" | ||
|
|
||
| managed_secrets="$(kubectl -n "$RELEASE_NAMESPACE" get secrets -l "${LABEL_KEY}=${LABEL_VALUE}" -o jsonpath='{range .items[*]}{.metadata.name}{"\n"}{end}')" | ||
|
|
||
| if [ -z "$managed_secrets" ]; then | ||
| log "no rotator-managed cluster secrets found in namespace ${RELEASE_NAMESPACE}" | ||
| exit 0 | ||
| fi | ||
|
|
||
| failure_count=0 | ||
|
|
||
| for secret_name in $managed_secrets; do | ||
| if ! reconcile_secret "$secret_name"; then | ||
| failure_count=$((failure_count + 1)) | ||
| fi | ||
| done | ||
|
|
||
| if [ "$failure_count" -gt 0 ]; then | ||
| log "${failure_count} secret reconciliation(s) failed" | ||
| exit 1 | ||
| fi | ||
|
|
||
| log "cluster credential rotation completed successfully" |
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is the stupid and simple approach to automatically onboarding and managing clusters but maybe we should consider building a full operator / controller for this? 🤔
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
But do we really need a runtime for this and do we want to built and distribute a custom container?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We could consider an inbetween custom container with simple go tooling but still keeping it as a cronjob instead of a constantly running pod?