Adds prometheus export point to the agent pod.

changelog.d/2975.added.md

@@ -0,0 +1 @@
+Add prometheus metrics to the mirrord-agent.

mirrord-schema.json

@@ -1,7 +1,7 @@
 {
   "$schema": "http://json-schema.org/draft-07/schema#",
   "title": "LayerFileConfig",
-  "description": "mirrord allows for a high degree of customization when it comes to which features you want to enable, and how they should function.\n\nAll of the configuration fields have a default value, so a minimal configuration would be no configuration at all.\n\nThe configuration supports templating using the [Tera](https://keats.github.io/tera/docs/) template engine. Currently we don't provide additional values to the context, if you have anything you want us to provide please let us know.\n\nTo use a configuration file in the CLI, use the `-f <CONFIG_PATH>` flag. Or if using VSCode Extension or JetBrains plugin, simply create a `.mirrord/mirrord.json` file or use the UI.\n\nTo help you get started, here are examples of a basic configuration file, and a complete configuration file containing all fields.\n\n### Basic `config.json` {#root-basic}\n\n```json { \"target\": \"pod/bear-pod\", \"feature\": { \"env\": true, \"fs\": \"read\", \"network\": true } } ```\n\n### Basic `config.json` with templating {#root-basic-templating}\n\n```json { \"target\": \"{{ get_env(name=\"TARGET\", default=\"pod/fallback\") }}\", \"feature\": { \"env\": true, \"fs\": \"read\", \"network\": true } } ```\n\n### Complete `config.json` {#root-complete}\n\nDon't use this example as a starting point, it's just here to show you all the available options. ```json { \"accept_invalid_certificates\": false, \"skip_processes\": \"ide-debugger\", \"target\": { \"path\": \"pod/bear-pod\", \"namespace\": \"default\" }, \"connect_tcp\": null, \"agent\": { \"log_level\": \"info\", \"json_log\": false, \"labels\": { \"user\": \"meow\" }, \"annotations\": { \"cats.io/inject\": \"enabled\" }, \"namespace\": \"default\", \"image\": \"ghcr.io/metalbear-co/mirrord:latest\", \"image_pull_policy\": \"IfNotPresent\", \"image_pull_secrets\": [ { \"secret-key\": \"secret\" } ], \"ttl\": 30, \"ephemeral\": false, \"communication_timeout\": 30, \"startup_timeout\": 360, \"network_interface\": \"eth0\", \"flush_connections\": true }, \"feature\": { \"env\": { \"include\": \"DATABASE_USER;PUBLIC_ENV\", \"exclude\": \"DATABASE_PASSWORD;SECRET_ENV\", \"override\": { \"DATABASE_CONNECTION\": \"db://localhost:7777/my-db\", \"LOCAL_BEAR\": \"panda\" }, \"mapping\": { \".+_TIMEOUT\": \"1000\" } }, \"fs\": { \"mode\": \"write\", \"read_write\": \".+\\\\.json\" , \"read_only\": [ \".+\\\\.yaml\", \".+important-file\\\\.txt\" ], \"local\": [ \".+\\\\.js\", \".+\\\\.mjs\" ] }, \"network\": { \"incoming\": { \"mode\": \"steal\", \"http_filter\": { \"header_filter\": \"host: api\\\\..+\" }, \"port_mapping\": [[ 7777, 8888 ]], \"ignore_localhost\": false, \"ignore_ports\": [9999, 10000] }, \"outgoing\": { \"tcp\": true, \"udp\": true, \"filter\": { \"local\": [\"tcp://1.1.1.0/24:1337\", \"1.1.5.0/24\", \"google.com\", \":53\"] }, \"ignore_localhost\": false, \"unix_streams\": \"bear.+\" }, \"dns\": { \"enabled\": true, \"filter\": { \"local\": [\"1.1.1.0/24:1337\", \"1.1.5.0/24\", \"google.com\"] } } }, \"copy_target\": { \"scale_down\": false } }, \"operator\": true, \"kubeconfig\": \"~/.kube/config\", \"sip_binaries\": \"bash\", \"telemetry\": true, \"kube_context\": \"my-cluster\" } ```\n\n# Options {#root-options}",
+  "description": "mirrord allows for a high degree of customization when it comes to which features you want to enable, and how they should function.\n\nAll of the configuration fields have a default value, so a minimal configuration would be no configuration at all.\n\nThe configuration supports templating using the [Tera](https://keats.github.io/tera/docs/) template engine. Currently we don't provide additional values to the context, if you have anything you want us to provide please let us know.\n\nTo use a configuration file in the CLI, use the `-f <CONFIG_PATH>` flag. Or if using VSCode Extension or JetBrains plugin, simply create a `.mirrord/mirrord.json` file or use the UI.\n\nTo help you get started, here are examples of a basic configuration file, and a complete configuration file containing all fields.\n\n### Basic `config.json` {#root-basic}\n\n```json { \"target\": \"pod/bear-pod\", \"feature\": { \"env\": true, \"fs\": \"read\", \"network\": true } } ```\n\n### Basic `config.json` with templating {#root-basic-templating}\n\n```json { \"target\": \"{{ get_env(name=\"TARGET\", default=\"pod/fallback\") }}\", \"feature\": { \"env\": true, \"fs\": \"read\", \"network\": true } } ```\n\n### Complete `config.json` {#root-complete}\n\nDon't use this example as a starting point, it's just here to show you all the available options. ```json { \"accept_invalid_certificates\": false, \"skip_processes\": \"ide-debugger\", \"target\": { \"path\": \"pod/bear-pod\", \"namespace\": \"default\" }, \"connect_tcp\": null, \"agent\": { \"log_level\": \"info\", \"json_log\": false, \"labels\": { \"user\": \"meow\" }, \"annotations\": { \"cats.io/inject\": \"enabled\" }, \"namespace\": \"default\", \"image\": \"ghcr.io/metalbear-co/mirrord:latest\", \"image_pull_policy\": \"IfNotPresent\", \"image_pull_secrets\": [ { \"secret-key\": \"secret\" } ], \"ttl\": 30, \"ephemeral\": false, \"communication_timeout\": 30, \"startup_timeout\": 360, \"network_interface\": \"eth0\", \"flush_connections\": true, \"metrics\": \"0.0.0.0:9000\", }, \"feature\": { \"env\": { \"include\": \"DATABASE_USER;PUBLIC_ENV\", \"exclude\": \"DATABASE_PASSWORD;SECRET_ENV\", \"override\": { \"DATABASE_CONNECTION\": \"db://localhost:7777/my-db\", \"LOCAL_BEAR\": \"panda\" }, \"mapping\": { \".+_TIMEOUT\": \"1000\" } }, \"fs\": { \"mode\": \"write\", \"read_write\": \".+\\\\.json\" , \"read_only\": [ \".+\\\\.yaml\", \".+important-file\\\\.txt\" ], \"local\": [ \".+\\\\.js\", \".+\\\\.mjs\" ] }, \"network\": { \"incoming\": { \"mode\": \"steal\", \"http_filter\": { \"header_filter\": \"host: api\\\\..+\" }, \"port_mapping\": [[ 7777, 8888 ]], \"ignore_localhost\": false, \"ignore_ports\": [9999, 10000] }, \"outgoing\": { \"tcp\": true, \"udp\": true, \"filter\": { \"local\": [\"tcp://1.1.1.0/24:1337\", \"1.1.5.0/24\", \"google.com\", \":53\"] }, \"ignore_localhost\": false, \"unix_streams\": \"bear.+\" }, \"dns\": { \"enabled\": true, \"filter\": { \"local\": [\"1.1.1.0/24:1337\", \"1.1.5.0/24\", \"google.com\"] } } }, \"copy_target\": { \"scale_down\": false } }, \"operator\": true, \"kubeconfig\": \"~/.kube/config\", \"sip_binaries\": \"bash\", \"telemetry\": true, \"kube_context\": \"my-cluster\" } ```\n\n# Options {#root-options}",
   "type": "object",
   "properties": {
     "accept_invalid_certificates": {
@@ -255,7 +255,7 @@
       "properties": {
         "annotations": {
           "title": "agent.annotations {#agent-annotations}",
-          "description": "Allows setting up custom annotations for the agent Job and Pod.\n\n```json { \"annotations\": { \"cats.io/inject\": \"enabled\" } } ```",
+          "description": "Allows setting up custom annotations for the agent Job and Pod.\n\n```json { \"annotations\": { \"cats.io/inject\": \"enabled\" \"prometheus.io/scrape\": \"true\", \"prometheus.io/port\": \"9000\" } } ```",
           "type": [
             "object",
             "null"
@@ -378,6 +378,14 @@
             "null"
           ]
         },
+        "metrics": {
+          "title": "agent.metrics {#agent-metrics}",
+          "description": "Enables prometheus metrics for the agent pod.\n\nYou might need to add annotations to the agent pod depending on how prometheus is configured to scrape for metrics.\n\n```json { \"metrics\": \"0.0.0.0:9000\" } ```",
+          "type": [
+            "string",
+            "null"
+          ]
+        },
         "namespace": {
           "title": "agent.namespace {#agent-namespace}",
           "description": "Namespace where the agent shall live. Note: Doesn't work with ephemeral containers. Defaults to the current kubernetes namespace.",

mirrord/agent/Cargo.toml

@@ -69,6 +69,8 @@ x509-parser = "0.16"
 rustls.workspace = true
 envy = "0.4"
 socket2.workspace = true
+prometheus = { version = "0.13", features = ["process"] }
+axum = { version = "0.7", features = ["macros"] }
 iptables = { git = "https://github.com/metalbear-co/rust-iptables.git", rev = "e66c7332e361df3c61a194f08eefe3f40763d624" }
 rawsocket = { git = "https://github.com/metalbear-co/rawsocket.git" }
 procfs = "0.17.0"
@@ -78,3 +80,4 @@ rstest.workspace = true
 mockall = "0.13"
 test_bin = "0.4"
 rcgen.workspace = true
+reqwest.workspace = true

mirrord/agent/README.md

@@ -6,3 +6,237 @@ Agent part of [mirrord](https://github.com/metalbear-co/mirrord) responsible for
 mirrord-agent is written in Rust for safety, low memory consumption and performance.
 
 mirrord-agent is distributed as a container image (currently only x86) that is published on [GitHub Packages publicly](https://github.com/metalbear-co/mirrord-agent/pkgs/container/mirrord-agent). 
+
+## Enabling prometheus metrics
+
+To start the metrics server, you'll need to add this config to your `mirrord.json`:
+
+```json
+{
+  "agent": {
+    "metrics": "0.0.0.0:9000",
+    "annotations": {
+      "prometheus.io/scrape": "true",
+      "prometheus.io/port": "9000"
+    }
+}
+```
+
+Remember to change the `port` in both `metrics` and `annotations`, they have to match,
+otherwise prometheus will try to scrape on `port: 80` or other commonly used ports.
+
+### Installing prometheus
+
+Run `kubectl apply -f {file-name}.yaml` on these sequences of `yaml` files and you should
+get prometheus running in your cluster. You can access the dashboard from your browser at
+`http://{cluster-ip}:30909`, if you're using minikube it might be
+`http://192.168.49.2:30909`.
+
+You'll get prometheus running under the `monitoring` namespace, but it'll be able to look
+into resources from all namespaces. The config in `configmap.yaml` sets prometheus to look
+at pods only, if you want to use it to scrape other stuff, check
+[this example](https://github.com/prometheus/prometheus/blob/main/documentation/examples/prometheus-kubernetes.yml).
+
+1. `create-namespace.yaml`
+
+```yaml
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: monitoring
+```
+
+2. `cluster-role.yaml`
+
+```yaml
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: prometheus
+rules:
+- apiGroups: [""]
+  resources:
+  - nodes
+  - services
+  - endpoints
+  - pods
+  verbs: ["get", "list", "watch"]
+- apiGroups:
+  - extensions
+  resources:
+  - ingresses
+  verbs: ["get", "list", "watch"]
+```
+
+3. `service-account.yaml`
+
+```yaml
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: prometheus
+  namespace: monitoring
+```
+
+4. `cluster-role-binding.yaml`
+
+```yaml
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: prometheus
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: prometheus
+subjects:
+- kind: ServiceAccount
+  name: prometheus
+  namespace: monitoring
+```
+
+5. `configmap.yaml`
+
+```yaml
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: prometheus-config
+  namespace: monitoring
+data:
+  prometheus.yml: |
+    # A scrape configuration for running Prometheus on a Kubernetes cluster.
+    # This uses separate scrape configs for cluster components (i.e. API server, node)
+    # and services to allow each to use different authentication configs.
+    #
+    # Kubernetes labels will be added as Prometheus labels on metrics via the
+    # `labelmap` relabeling action.
+    #
+    # If you are using Kubernetes 1.7.2 or earlier, please take note of the comments
+    # for the kubernetes-cadvisor job; you will need to edit or remove this job.
+
+    # Keep at most 100 sets of details of targets dropped by relabeling.
+    # This information is used to display in the UI for troubleshooting.
+    global:
+      keep_dropped_targets: 100
+
+    # Scrape config for API servers.
+    #
+    # Kubernetes exposes API servers as endpoints to the default/kubernetes
+    # service so this uses `endpoints` role and uses relabelling to only keep
+    # the endpoints associated with the default/kubernetes service using the
+    # default named port `https`. This works for single API server deployments as
+    # well as HA API server deployments.
+    scrape_configs:
+      # Example scrape config for pods
+      #
+      # The relabeling allows the actual pod scrape to be configured
+      # for all the declared ports (or port-free target if none is declared)
+      # or only some ports.
+      - job_name: "kubernetes-pods"
+
+        kubernetes_sd_configs:
+          - role: pod
+
+        relabel_configs:
+          # Example relabel to scrape only pods that have
+          # "example.io/should_be_scraped = true" annotation.
+          #  - source_labels: [__meta_kubernetes_pod_annotation_example_io_should_be_scraped]
+          #    action: keep
+          #    regex: true
+          #
+          # Example relabel to customize metric path based on pod
+          # "example.io/metric_path = <metric path>" annotation.
+          #  - source_labels: [__meta_kubernetes_pod_annotation_example_io_metric_path]
+          #    action: replace
+          #    target_label: __metrics_path__
+          #    regex: (.+)
+          #
+          # Example relabel to scrape only single, desired port for the pod
+          # based on pod "example.io/scrape_port = <port>" annotation.
+          - source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_port]
+            action: replace
+            regex: ([^:]+)(?::\d+)?;(\d+)
+            replacement: $1:$2
+            target_label: __address__
+          - action: labelmap
+            regex: __meta_kubernetes_pod_label_(.+)
+          - source_labels: [__meta_kubernetes_namespace]
+            action: replace
+            target_label: namespace
+          - source_labels: [__meta_kubernetes_pod_name]
+            action: replace
+            target_label: pod
+```
+
+- If you make any changes to the 5-configmap.yaml file, remember to `kubectl apply` it
+ **before** restarting the `prometheus` deployment.
+
+6. `deployment.yaml`
+
+```yaml
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: prometheus
+  namespace: monitoring
+  labels:
+    app: prometheus
+spec:
+  replicas: 1
+  strategy:
+    rollingUpdate:
+      maxSurge: 1
+      maxUnavailable: 1
+    type: RollingUpdate
+  selector:
+    matchLabels:
+      app: prometheus
+  template:
+    metadata:
+      labels:
+        app: prometheus
+      annotations:
+        prometheus.io/scrape: "true"
+        prometheus.io/port: "9090"
+    spec:
+      serviceAccountName: prometheus
+      containers:
+      - name: prometheus
+        image: prom/prometheus
+        args:
+          - '--config.file=/etc/prometheus/prometheus.yml'
+        ports:
+        - name: web
+          containerPort: 9090
+        volumeMounts:
+        - name: prometheus-config-volume
+          mountPath: /etc/prometheus
+      restartPolicy: Always
+      volumes:
+      - name: prometheus-config-volume
+        configMap:
+            defaultMode: 420
+            name: prometheus-config
+```
+
+7. `service.yaml`
+
+```yaml
+apiVersion: v1
+kind: Service
+metadata:
+    name: prometheus-service
+    namespace: monitoring
+    annotations:
+        prometheus.io/scrape: 'true'
+        prometheus.io/port:   '9090'
+spec:
+    selector:
+        app: prometheus
+    type: NodePort
+    ports:
+    - port: 8080
+      targetPort: 9090
+      nodePort: 30909
+```

mirrord/agent/src/cli.rs

@@ -1,8 +1,11 @@
 #![deny(missing_docs)]
 
+use std::net::SocketAddr;
+
 use clap::{Parser, Subcommand};
 use mirrord_protocol::{
-    MeshVendor, AGENT_IPV6_ENV, AGENT_NETWORK_INTERFACE_ENV, AGENT_OPERATOR_CERT_ENV,
+    MeshVendor, AGENT_IPV6_ENV, AGENT_METRICS_ENV, AGENT_NETWORK_INTERFACE_ENV,
+    AGENT_OPERATOR_CERT_ENV,
 };
 
 const DEFAULT_RUNTIME: &str = "containerd";
@@ -28,6 +31,10 @@ pub struct Args {
     #[arg(short = 'i', long, env = AGENT_NETWORK_INTERFACE_ENV)]
     pub network_interface: Option<String>,
 
+    /// Controls whether metrics are enabled, and the address to set up the metrics server.
+    #[arg(long, env = AGENT_METRICS_ENV)]
+    pub metrics: Option<SocketAddr>,
+
     /// Return an error after accepting the first client connection, in order to test agent error
     /// cleanup.
     ///

mirrord/agent/src/client_connection.rs

@@ -208,7 +208,7 @@ enum ConnectionFramed {
 
 #[cfg(test)]
 mod test {
-    use std::sync::Arc;
+    use std::sync::{Arc, Once};
 
     use futures::StreamExt;
     use mirrord_protocol::ClientCodec;
@@ -220,10 +220,19 @@ mod test {
 
     use super::*;
 
+    static CRYPTO_PROVIDER: Once = Once::new();
+
     /// Verifies that [`AgentTlsConnector`] correctly accepts a
     /// connection from a server using the provided certificate.
     #[tokio::test]
     async fn agent_tls_connector_valid_cert() {
+        CRYPTO_PROVIDER.call_once(|| {
+            rustls::crypto::CryptoProvider::install_default(
+                rustls::crypto::aws_lc_rs::default_provider(),
+            )
+            .expect("Failed to install crypto provider")
+        });
+
         let cert = rcgen::generate_simple_self_signed(vec!["operator".to_string()]).unwrap();
         let cert_bytes = cert.cert.der();
         let key_bytes = cert.key_pair.serialize_der();
@@ -269,6 +278,13 @@ mod test {
     /// connection from a server using some other certificate.
     #[tokio::test]
     async fn agent_tls_connector_invalid_cert() {
+        CRYPTO_PROVIDER.call_once(|| {
+            rustls::crypto::CryptoProvider::install_default(
+                rustls::crypto::aws_lc_rs::default_provider(),
+            )
+            .expect("Failed to install crypto provider")
+        });
+
         let server_cert = rcgen::generate_simple_self_signed(vec!["operator".to_string()]).unwrap();
         let cert_bytes = server_cert.cert.der();
         let key_bytes = server_cert.key_pair.serialize_der();

mirrord/agent/src/container_handle.rs

@@ -1,7 +1,7 @@
 use std::{collections::HashMap, sync::Arc};
 
 use crate::{
-    error::Result,
+    error::AgentResult,
     runtime::{Container, ContainerInfo, ContainerRuntime},
 };
 
@@ -22,7 +22,7 @@ pub(crate) struct ContainerHandle(Arc<Inner>);
 impl ContainerHandle {
     /// Retrieve info about the container and initialize this struct.
     #[tracing::instrument(level = "trace")]
-    pub(crate) async fn new(container: Container) -> Result<Self> {
+    pub(crate) async fn new(container: Container) -> AgentResult<Self> {
         let ContainerInfo { pid, env: raw_env } = container.get_info().await?;
 
         let inner = Inner { pid, raw_env };