Merge pull request 'improve-bootstrap-visibility' (#29) from improve-bootstrap-visibility into master

Reviewed-on: #29

CLAUDE.md

@@ -61,10 +61,31 @@ Do not run `nix flake update`. Should only be done manually by user.
 ### Development Environment
 
 ```bash
-# Enter development shell (provides ansible, python3)
+# Enter development shell
 nix develop
 ```
 
+The devshell provides: `ansible`, `tofu` (OpenTofu), `vault` (OpenBao CLI), `create-host`, and `homelab-deploy`.
+
+**Important:** When suggesting commands that use devshell tools, always use `nix develop -c <command>` syntax rather than assuming the user is already in a devshell. For example:
+```bash
+# Good - works regardless of current shell
+nix develop -c tofu plan
+
+# Avoid - requires user to be in devshell
+tofu plan
+```
+
+**OpenTofu:** Use the `-chdir` option instead of `cd` when running tofu commands in subdirectories:
+```bash
+# Good - uses -chdir option
+nix develop -c tofu -chdir=terraform plan
+nix develop -c tofu -chdir=terraform/vault apply
+
+# Avoid - changing directories
+cd terraform && tofu plan
+```
+
 ### Secrets Management
 
 Secrets are managed by OpenBao (Vault) using AppRole authentication. Most hosts use the
@@ -140,11 +161,27 @@ The **lab-monitoring** MCP server can query logs from Loki. All hosts ship syste
 
 - `host` - Hostname (e.g., `ns1`, `ns2`, `monitoring01`, `ha1`). Use this label, not `hostname`.
 - `systemd_unit` - Systemd unit name (e.g., `nsd.service`, `prometheus.service`, `nixos-upgrade.service`)
-- `job` - Either `systemd-journal` (most logs) or `varlog` (file-based logs like caddy access logs)
+- `job` - Either `systemd-journal` (most logs), `varlog` (file-based logs), or `bootstrap` (VM bootstrap logs)
 - `filename` - For `varlog` job, the log file path (e.g., `/var/log/caddy/nix-cache.log`)
 
 Journal log entries are JSON-formatted with the actual log message in the `MESSAGE` field. Other useful fields include `PRIORITY` and `SYSLOG_IDENTIFIER`.
 
+**Bootstrap Logs:**
+
+VMs provisioned from template2 send bootstrap progress directly to Loki via curl (before promtail is available). These logs use `job="bootstrap"` with additional labels:
+
+- `host` - Target hostname
+- `branch` - Git branch being deployed
+- `stage` - Bootstrap stage: `starting`, `network_ok`, `vault_ok`/`vault_skip`/`vault_warn`, `building`, `success`, `failed`
+
+Query bootstrap status:
+```
+{job="bootstrap"}                              # All bootstrap logs
+{job="bootstrap", host="testvm01"}             # Specific host
+{job="bootstrap", stage="failed"}              # All failures
+{job="bootstrap", stage=~"building|success"}   # Track build progress
+```
+
 **Example LogQL queries:**
 ```
 # Logs from a specific service on a host

hosts/template2/bootstrap.nix

@@ -6,22 +6,72 @@ let
     text = ''
       set -euo pipefail
 
+      LOKI_URL="http://monitoring01.home.2rjus.net:3100/loki/api/v1/push"
+
+      # Send a log entry to Loki with bootstrap status
+      # Usage: log_to_loki <stage> <message>
+      # Fails silently if Loki is unreachable
+      log_to_loki() {
+        local stage="$1"
+        local message="$2"
+        local timestamp_ns
+        timestamp_ns="$(date +%s)000000000"
+
+        local payload
+        payload=$(jq -n \
+          --arg host "$HOSTNAME" \
+          --arg stage "$stage" \
+          --arg branch "''${BRANCH:-master}" \
+          --arg ts "$timestamp_ns" \
+          --arg msg "$message" \
+          '{
+            streams: [{
+              stream: {
+                job: "bootstrap",
+                host: $host,
+                stage: $stage,
+                branch: $branch
+              },
+              values: [[$ts, $msg]]
+            }]
+          }')
+
+        curl -s --connect-timeout 2 --max-time 5 \
+          -X POST \
+          -H "Content-Type: application/json" \
+          -d "$payload" \
+          "$LOKI_URL" >/dev/null 2>&1 || true
+      }
+
+      echo "================================================================================"
+      echo "                     NIXOS BOOTSTRAP IN PROGRESS"
+      echo "================================================================================"
+      echo ""
+
       # Read hostname set by cloud-init (from Terraform VM name via user-data)
       # Cloud-init sets the system hostname from user-data.txt, so we read it from hostnamectl
       HOSTNAME=$(hostnamectl hostname)
-      echo "DEBUG: Hostname from hostnamectl: '$HOSTNAME'"
+      # Read git branch from environment, default to master
+      BRANCH="''${NIXOS_FLAKE_BRANCH:-master}"
 
+      echo "Hostname: $HOSTNAME"
+      echo ""
       echo "Starting NixOS bootstrap for host: $HOSTNAME"
+
+      log_to_loki "starting" "Bootstrap starting for $HOSTNAME (branch: $BRANCH)"
+
       echo "Waiting for network connectivity..."
 
       # Verify we can reach the git server via HTTPS (doesn't respond to ping)
       if ! curl -s --connect-timeout 5 --max-time 10 https://git.t-juice.club >/dev/null 2>&1; then
         echo "ERROR: Cannot reach git.t-juice.club via HTTPS"
         echo "Check network configuration and DNS settings"
+        log_to_loki "failed" "Network check failed - cannot reach git.t-juice.club"
         exit 1
       fi
 
       echo "Network connectivity confirmed"
+      log_to_loki "network_ok" "Network connectivity confirmed"
 
       # Unwrap Vault token and store AppRole credentials (if provided)
       if [ -n "''${VAULT_WRAPPED_TOKEN:-}" ]; then
@@ -50,6 +100,7 @@ let
           chmod 600 /var/lib/vault/approle/secret-id
 
           echo "Vault credentials unwrapped and stored successfully"
+          log_to_loki "vault_ok" "Vault credentials unwrapped and stored"
         else
           echo "WARNING: Failed to unwrap Vault token"
           if [ -n "$UNWRAP_RESPONSE" ]; then
@@ -63,17 +114,17 @@ let
           echo "To regenerate token, run: create-host --hostname $HOSTNAME --force"
           echo ""
           echo "Vault secrets will not be available, but continuing bootstrap..."
+          log_to_loki "vault_warn" "Failed to unwrap Vault token - continuing without secrets"
         fi
       else
         echo "No Vault wrapped token provided (VAULT_WRAPPED_TOKEN not set)"
         echo "Skipping Vault credential setup"
+        log_to_loki "vault_skip" "No Vault token provided - skipping credential setup"
       fi
 
       echo "Fetching and building NixOS configuration from flake..."
-
-      # Read git branch from environment, default to master
-      BRANCH="''${NIXOS_FLAKE_BRANCH:-master}"
       echo "Using git branch: $BRANCH"
+      log_to_loki "building" "Starting nixos-rebuild boot"
 
       # Build and activate the host-specific configuration
       FLAKE_URL="git+https://git.t-juice.club/torjus/nixos-servers.git?ref=$BRANCH#''${HOSTNAME}"
@@ -81,18 +132,30 @@ let
       if nixos-rebuild boot --flake "$FLAKE_URL"; then
         echo "Successfully built configuration for $HOSTNAME"
         echo "Rebooting into new configuration..."
+        log_to_loki "success" "Build successful - rebooting into new configuration"
         sleep 2
         systemctl reboot
       else
         echo "ERROR: nixos-rebuild failed for $HOSTNAME"
         echo "Check that flake has configuration for this hostname"
         echo "Manual intervention required - system will not reboot"
+        log_to_loki "failed" "nixos-rebuild failed - manual intervention required"
         exit 1
       fi
     '';
   };
 in
 {
+  # Custom greeting line to indicate this is a bootstrap image
+  services.getty.greetingLine = lib.mkForce ''
+    ================================================================================
+                          BOOTSTRAP IMAGE - NixOS \V (\l)
+    ================================================================================
+
+    Bootstrap service is running. Logs are displayed on tty1.
+    Check status: journalctl -fu nixos-bootstrap
+  '';
+
   systemd.services."nixos-bootstrap" = {
     description = "Bootstrap NixOS configuration from flake on first boot";
 
@@ -107,12 +170,12 @@ in
     serviceConfig = {
       Type = "oneshot";
       RemainAfterExit = true;
-      ExecStart = "${bootstrap-script}/bin/nixos-bootstrap";
+      ExecStart = lib.getExe bootstrap-script;
 
       # Read environment variables from cloud-init (set by cloud-init write_files)
       EnvironmentFile = "-/run/cloud-init-env";
 
-      # Logging to journald
+      # Log to journal and console
       StandardOutput = "journal+console";
       StandardError = "journal+console";
     };

playbooks/build-and-deploy-template.yml

@@ -99,3 +99,48 @@
     - name: Display success message
       ansible.builtin.debug:
         msg: "Template VM {{ template_vmid }} created successfully on {{ storage }}"
+
+- name: Update Terraform template name
+  hosts: localhost
+  gather_facts: false
+
+  vars:
+    terraform_dir: "{{ playbook_dir }}/../terraform"
+
+  tasks:
+    - name: Get image filename from earlier play
+      ansible.builtin.set_fact:
+        image_filename: "{{ hostvars['localhost']['image_filename'] }}"
+
+    - name: Extract template name from image filename
+      ansible.builtin.set_fact:
+        new_template_name: "{{ image_filename | regex_replace('\\.vma\\.zst$', '') | regex_replace('^vzdump-qemu-', '') }}"
+
+    - name: Read current Terraform variables file
+      ansible.builtin.slurp:
+        src: "{{ terraform_dir }}/variables.tf"
+      register: variables_tf_content
+
+    - name: Extract current template name from variables.tf
+      ansible.builtin.set_fact:
+        current_template_name: "{{ (variables_tf_content.content | b64decode) | regex_search('variable \"default_template_name\"[^}]+default\\s*=\\s*\"([^\"]+)\"', '\\1') | first }}"
+
+    - name: Check if template name has changed
+      ansible.builtin.set_fact:
+        template_name_changed: "{{ current_template_name != new_template_name }}"
+
+    - name: Display template name status
+      ansible.builtin.debug:
+        msg: "Template name: {{ current_template_name }} -> {{ new_template_name }} ({{ 'changed' if template_name_changed else 'unchanged' }})"
+
+    - name: Update default_template_name in variables.tf
+      ansible.builtin.replace:
+        path: "{{ terraform_dir }}/variables.tf"
+        regexp: '(variable "default_template_name"[^}]+default\s*=\s*)"[^"]+"'
+        replace: '\1"{{ new_template_name }}"'
+      when: template_name_changed
+
+    - name: Display update result
+      ansible.builtin.debug:
+        msg: "Updated terraform/variables.tf with new template name: {{ new_template_name }}"
+      when: template_name_changed

terraform/variables.tf

@@ -33,7 +33,7 @@ variable "default_target_node" {
 variable "default_template_name" {
   description = "Default template VM name to clone from"
   type        = string
-  default     = "nixos-25.11.20260131.41e216c"
+  default     = "nixos-25.11.20260203.e576e3c"
 }
 
 variable "default_ssh_public_key" {

terraform/vms.tf

@@ -39,10 +39,12 @@ locals {
       flake_branch = "vault-setup" # Bootstrap from this branch instead of master
     }
     "testvm01" = {
-      ip        = "10.69.13.20/24"
+      ip                  = "10.69.13.20/24"
-      cpu_cores = 2
+      cpu_cores           = 2
-      memory    = 2048
+      memory              = 2048
-      disk_size = "20G"
+      disk_size           = "20G"
+      flake_branch        = "improve-bootstrap-visibility"
+      vault_wrapped_token = "s.l5q88wzXfEcr5SMDHmO6o96b"
     }
     "testvm02" = {
       ip        = "10.69.13.21/24"