Add DRE tool command to fetch "untrusted" metrics from metrics-canister

Cargo.Bazel.lock

@@ -1,5 +1,5 @@
 {
-  "checksum": "f3c6195e6654db56b4c772ec401cc5689f9b0fa6466d7c0fc6543302e0774321",
+  "checksum": "21560a1c6cf60a5df7d6ab8a9c786242587fcc03a6171df0fc1574fc13c8911c",
   "crates": {
     "actix-codec 0.5.2": {
       "name": "actix-codec",
@@ -28369,29 +28369,35 @@
         ],
         "crate_features": {
           "common": [
-            "elf",
-            "errno",
             "general",
             "ioctl",
             "no_std"
           ],
           "selects": {
             "aarch64-unknown-linux-gnu": [
+              "elf",
+              "errno",
               "prctl",
               "std",
               "system"
             ],
             "arm-unknown-linux-gnueabi": [
+              "elf",
+              "errno",
               "prctl",
               "std",
               "system"
             ],
             "armv7-unknown-linux-gnueabi": [
+              "elf",
+              "errno",
               "prctl",
               "std",
               "system"
             ],
             "i686-unknown-linux-gnu": [
+              "elf",
+              "errno",
               "prctl",
               "std",
               "system"
@@ -28407,6 +28413,8 @@
               "system"
             ],
             "x86_64-unknown-linux-gnu": [
+              "elf",
+              "errno",
               "prctl",
               "std",
               "system"
@@ -45604,8 +45612,18 @@
       "name": "trustworthy-node-metrics",
       "version": "0.4.3",
       "repository": null,
-      "targets": [],
-      "library_target_name": null,
+      "targets": [
+        {
+          "Library": {
+            "crate_name": "trustworthy_node_metrics",
+            "crate_root": "src/lib.rs",
+            "srcs": [
+              "**/*.rs"
+            ]
+          }
+        }
+      ],
+      "library_target_name": "trustworthy_node_metrics",
       "common_attrs": {
         "compile_data_glob": [
           "**"

Cargo.toml

@@ -73,6 +73,7 @@ cryptoki = "0.3.1"
 csv = "1.3.0"
 custom_error = "1.9.2"
 decentralization = { path = "rs/decentralization" }
+trustworthy-node-metrics = { path = "rs/dre-canisters/trustworthy-node-metrics/src/trustworthy-node-metrics" }
 derive_builder = "0.20.0"
 derive_more = "0.99.18"
 dialoguer = "0.11.0"

docs/trustworthy-metrics/trustworthy-metrics.md

@@ -92,7 +92,7 @@ You can obtain the DRE tool by following the instructions from [getting started]
 To test out the command you can run the following command
 
 ```bash
-dre <auth-params> trustworthy-metrics <wallet-canister-id> <start-at-timestamp> [<subnet-id>...]
+dre <auth-params> node-metrics --trustworthy --wallet <wallet-canister-id> <start-at-timestamp> [<subnet-id>...]
 ```
 
 ??? tip "Explanation of the arguments"
@@ -149,12 +149,12 @@ Authentication with a private key is recommended, since it allows for more paral
 Here are some real-world examples of how metrics can be retrieved:
 
 ```bash
-dre --private-key-pem identity.pem trustworthy-metrics nanx4-baaaa-aaaap-qb4sq-cai 0 > data.json
+dre --private-key-pem identity.pem node-metrics --trustworthy --wallet nanx4-baaaa-aaaap-qb4sq-cai 0 > data.json
 ```
 
 Or with an HSM:
 ```bash
-dre --hsm-slot 0 --hsm-key-id 0 --hsm-pin "<pin>" trustworthy-metrics nanx4-baaaa-aaaap-qb4sq-cai 0 > data.json
+dre --hsm-slot 0 --hsm-key-id 0 --hsm-pin "<pin>" node-metrics --trustworthy --wallet nanx4-baaaa-aaaap-qb4sq-cai 0 > data.json
 ```
 
 You can check some examples of the analytics possible with the IC Mainnet data in the following [Jupyter Notebook](./TrustworthyMetricsAnalytics.ipynb)

docs/trustworthy-metrics/untrusted-metrics.md

@@ -0,0 +1,62 @@
+
+# Get untrusted metrics from Node Metrics canister
+
+## Introduction
+
+Untrusted Node Metrics retrieval offers an alternative approach to accessing node performance data, relying on a canister that collects these metrics instead of quering the management canister of each subnet directly.
+
+This method allows users to fetch node metrics dating back to May 18, 2024, providing an historical view compared to the trustworthy method, which only offers data from the past month.
+
+The key drawback of quering untrusted metrics is that it introduces an intermediary, the canister responsible for data aggregation, which should NOT be considered trustworthy.
+
+Despite these concerns, the extended temporal coverage can be valuable for certain analytical purposes. Additionally, querying the node metrics canister is cheaper because it allows for a query call instead of an update call and does not require a wallet canister.
+
+This entire process is shown in the following diagram:
+
+```mermaid
+%%{init: {'theme':'forest'}}%%
+graph TD
+    subgraph "Subnet 1"
+        S1["Consensus"] -->|Produces Trustworthy Data| M1["Management Canister 1"] --> M4["Node Metrics Canister"]
+    end
+    subgraph "Subnet 2"
+        S2["Consensus"] -->|Produces Trustworthy Data| M2["Management Canister 2"]
+    end
+    subgraph "Subnet 3"
+        S3["Consensus"] -->|Produces Trustworthy Data| M3["Management Canister 3"]
+    end
+    M2 --> M4
+    M3 --> M4
+    M4 --> DRE["DRE tool (open source)"]
+    DRE --> User
+    User --> |Analyze & Process Data| F["Node Metrics"]
+
+
+    style S1 fill:#f9f,stroke:#333,stroke-width:2px
+    style S2 fill:#f9f,stroke:#333,stroke-width:2px
+    style S3 fill:#f9f,stroke:#333,stroke-width:2px
+    style DRE fill:#ff9,stroke:#333,stroke-width:2px
+    style F fill:#9ff,stroke:#333,stroke-width:2px
+```
+
+### Using the cli
+
+You can obtain the DRE tool by following the instructions from [getting started](../getting-started.md)
+
+To test out the command you can run the following command
+
+```bash
+dre node-metrics <start-at-timestamp> [<subnet-id>...]
+```
+
+??? tip "Explanation of the arguments"
+    3. `start-at-timestamp` - used for filtering the output. To get all metrics, provide 0
+    4. `subnet-id` - subnets to query, if empty will provide metrics for all subnets
+
+# Example use
+
+Here are some real-world examples of how metrics can be retrieved:
+
+```bash
+dre node-metrics 0 > data.json
+```

rs/cli/BUILD.bazel

@@ -7,6 +7,7 @@ DEPS = [
     "//rs/decentralization",
     "//rs/ic-management-types",
     "//rs/ic-management-backend:ic-management-backend-lib",
+    "//rs/dre-canisters/trustworthy-node-metrics/src/trustworthy-node-metrics:trustworthy_node_metrics"
 ]
 
 package(default_visibility = ["//visibility:public"])

rs/cli/src/commands/mod.rs

@@ -10,11 +10,11 @@ use get::Get;
 use heal::Heal;
 use hostos::HostOsCmd;
 use ic_management_types::{MinNakamotoCoefficients, Network, NodeFeature};
+use node_metrics::NodeMetrics;
 use nodes::Nodes;
 use proposals::Proposals;
 use propose::Propose;
 use registry::Registry;
-use trustworthy_metrics::TrustworthyMetrics;
 use update_unassigned_nodes::UpdateUnassignedNodes;
 use upgrade::Upgrade;
 use url::Url;
@@ -30,12 +30,12 @@ mod firewall;
 mod get;
 mod heal;
 pub mod hostos;
+mod node_metrics;
 mod nodes;
 mod proposals;
 mod propose;
 mod registry;
 mod subnet;
-mod trustworthy_metrics;
 mod update_unassigned_nodes;
 pub mod upgrade;
 mod version;
@@ -123,6 +123,9 @@ pub enum Subcommands {
     /// Manage versions
     Version(VersionCmd),
 
+    /// Fetch node metrics stats
+    NodeMetrics(NodeMetrics),
+
     /// Manage hostos versions
     HostOs(HostOsCmd),
 
@@ -135,9 +138,6 @@ pub enum Subcommands {
     /// Vote on our proposals
     Vote(Vote),
 
-    /// Trustworthy Metrics
-    TrustworthyMetrics(TrustworthyMetrics),
-
     /// Registry inspection (dump) operations
     Registry(Registry),
 
@@ -251,12 +251,12 @@ impl ExecutableCommand for Args {
             Subcommands::Nodes(c) => c.require_ic_admin(),
             Subcommands::ApiBoundaryNodes(c) => c.require_ic_admin(),
             Subcommands::Vote(c) => c.require_ic_admin(),
-            Subcommands::TrustworthyMetrics(c) => c.require_ic_admin(),
             Subcommands::Registry(c) => c.require_ic_admin(),
             Subcommands::Firewall(c) => c.require_ic_admin(),
             Subcommands::Upgrade(c) => c.require_ic_admin(),
             Subcommands::Proposals(c) => c.require_ic_admin(),
             Subcommands::Completions(c) => c.require_ic_admin(),
+            Subcommands::NodeMetrics(c) => c.require_ic_admin(),
         }
     }
 
@@ -273,12 +273,12 @@ impl ExecutableCommand for Args {
             Subcommands::Nodes(c) => c.execute(ctx).await,
             Subcommands::ApiBoundaryNodes(c) => c.execute(ctx).await,
             Subcommands::Vote(c) => c.execute(ctx).await,
-            Subcommands::TrustworthyMetrics(c) => c.execute(ctx).await,
             Subcommands::Registry(c) => c.execute(ctx).await,
             Subcommands::Firewall(c) => c.execute(ctx).await,
             Subcommands::Upgrade(c) => c.execute(ctx).await,
             Subcommands::Proposals(c) => c.execute(ctx).await,
             Subcommands::Completions(c) => c.execute(ctx).await,
+            Subcommands::NodeMetrics(c) => c.execute(ctx).await,
         }
     }
 
@@ -295,12 +295,12 @@ impl ExecutableCommand for Args {
             Subcommands::Nodes(c) => c.validate(cmd),
             Subcommands::ApiBoundaryNodes(c) => c.validate(cmd),
             Subcommands::Vote(c) => c.validate(cmd),
-            Subcommands::TrustworthyMetrics(c) => c.validate(cmd),
             Subcommands::Registry(c) => c.validate(cmd),
             Subcommands::Firewall(c) => c.validate(cmd),
             Subcommands::Upgrade(c) => c.validate(cmd),
             Subcommands::Proposals(c) => c.validate(cmd),
             Subcommands::Completions(c) => c.validate(cmd),
+            Subcommands::NodeMetrics(c) => c.validate(cmd),
         }
     }
 }