Merge pull request 'removed netcat dependency with /dev/tcp' (#2) from bug/fix-missing-dependencies-k8s-bootstrap into main

Reviewed-on: #2

.gitea/workflows/deploy-k8s-bootstrap.yaml

@@ -85,31 +85,3 @@ jobs:
           cloud-url: ${{ secrets.PULUMI_BACKEND_URL }}
         env:
           PULUMI_CONFIG_PASSPHRASE: ${{ secrets.PULUMI_CONFIG_PASSPHRASE }}
-
-      # Propagate kubeconfig to the downstream stacks so their next deploy picks it up
-      - name: Propagate kubeconfig to k8s-infra
-        run: |
-          echo "${{ secrets.K8S_INFRA_DEV_YAML }}" | base64 -d > k8s-infra/Pulumi.dev.yaml
-          cd k8s-infra && npm install
-          KUBECONFIG=$(cd ../k8s-bootstrap && pulumi stack output kubeconfig --show-secrets \
-            --cloud-url "${{ secrets.PULUMI_BACKEND_URL }}" -s dev)
-          pulumi config set --secret kubeconfig "$KUBECONFIG" \
-            --cloud-url "${{ secrets.PULUMI_BACKEND_URL }}" -s dev
-          # Re-encode updated config for the secret (update manually in Gitea after first run)
-          base64 -w 0 Pulumi.dev.yaml
-        working-directory: .
-        env:
-          PULUMI_CONFIG_PASSPHRASE: ${{ secrets.PULUMI_CONFIG_PASSPHRASE }}
-
-      - name: Propagate kubeconfig to k8s-apps
-        run: |
-          echo "${{ secrets.K8S_APPS_DEV_YAML }}" | base64 -d > k8s-apps/Pulumi.dev.yaml
-          cd k8s-apps && npm install
-          KUBECONFIG=$(cd ../k8s-bootstrap && pulumi stack output kubeconfig --show-secrets \
-            --cloud-url "${{ secrets.PULUMI_BACKEND_URL }}" -s dev)
-          pulumi config set --secret kubeconfig "$KUBECONFIG" \
-            --cloud-url "${{ secrets.PULUMI_BACKEND_URL }}" -s dev
-          base64 -w 0 Pulumi.dev.yaml
-        working-directory: .
-        env:
-          PULUMI_CONFIG_PASSPHRASE: ${{ secrets.PULUMI_CONFIG_PASSPHRASE }}

k8s-bootstrap/CLAUDE.md

@@ -5,7 +5,7 @@ Bootstraps a k3s cluster on the 5 Proxmox VMs created by `proxmox-infra`. Starts
 ## How it works
 
 1. Starts all 5 VMs via `POST /api2/json/nodes/{node}/qemu/{vmid}/status/start`
-2. Waits for port 22 to open on each VM (`nc -z`)
+2. Waits for port 22 to open on each VM (bash `/dev/tcp`)
 3. Installs k3s on `k3s-master-1` with `--cluster-init --tls-san <master1Ip>`
 4. Joins `k3s-master-2` and `k3s-master-3` as embedded etcd nodes
 5. Joins `k3s-worker-1` and `k3s-worker-2` as agent nodes
@@ -18,12 +18,6 @@ VM IDs and the CI runner SSH private key are read automatically from the `proxmo
 Run in this directory after `pulumi stack init dev`:
 
 ```bash
-# Same as proxmox-infra
-pulumi config set --secret pve1Endpoint "https://192.168.1.x:8006"
-pulumi config set --secret pve1ApiToken "root@pam!pulumi=<token>"
-pulumi config set --secret pve2Endpoint "https://192.168.1.y:8006"
-pulumi config set --secret pve2ApiToken "root@pam!pulumi=<token>"
-
 # Pre-shared k3s token — any strong random string
 pulumi config set --secret k3sToken "$(openssl rand -hex 32)"
 
@@ -35,6 +29,14 @@ pulumi config set worker1Ip "192.168.1.x"
 pulumi config set worker2Ip "192.168.1.x"
 ```
 
+Proxmox credentials (`pve1Endpoint`, `pve1ApiToken`, `pve2Endpoint`, `pve2ApiToken`) are read automatically from the `proxmox-infra` stack outputs via StackReference — do **not** set them here.
+
+## Deployment order
+
+`proxmox-infra` must be deployed **before** k8s-bootstrap. The proxmox-infra stack exports the Proxmox credentials and VM IDs that k8s-bootstrap reads via StackReference. If proxmox-infra hasn't been re-deployed after its latest code changes, those outputs won't exist and k8s-bootstrap will fail.
+
+In CI, both workflows run as previews on PRs (no deploy). Trigger `workflow_dispatch` on `deploy-proxmox-infra` first, then run k8s-bootstrap.
+
 After setting config, re-encode `Pulumi.dev.yaml` and update the Gitea secret `K8S_BOOTSTRAP_DEV_YAML`:
 ```bash
 base64 -w 0 Pulumi.dev.yaml

k8s-bootstrap/index.ts

@@ -33,14 +33,18 @@ const worker2Ip = config.require("worker2Ip");
 const k3sToken = config.requireSecret("k3sToken");
 
 // VM IDs and CI runner SSH key — read from proxmox-infra stack outputs
-const vmIdsOutput = infraRef.requireOutput("vmIds") as pulumi.Output<Record<string, number>>;
-const ciRunnerPrivateKey = infraRef.requireOutput("ciRunnerPrivateKey") as pulumi.Output<string>;
+const vmIdsOutput = infraRef.requireOutput("vmIds") as pulumi.Output<
+  Record<string, number>
+>;
+const ciRunnerPrivateKey = infraRef.requireOutput(
+  "ciRunnerPrivateKey",
+) as pulumi.Output<string>;
 
-const master1VmId = vmIdsOutput.apply(ids => String(ids.master1));
-const master2VmId = vmIdsOutput.apply(ids => String(ids.master2));
-const master3VmId = vmIdsOutput.apply(ids => String(ids.master3));
-const worker1VmId = vmIdsOutput.apply(ids => String(ids.worker1));
-const worker2VmId = vmIdsOutput.apply(ids => String(ids.worker2));
+const master1VmId = vmIdsOutput.apply((ids) => String(ids.master1));
+const master2VmId = vmIdsOutput.apply((ids) => String(ids.master2));
+const master3VmId = vmIdsOutput.apply((ids) => String(ids.master3));
+const worker1VmId = vmIdsOutput.apply((ids) => String(ids.worker1));
+const worker2VmId = vmIdsOutput.apply((ids) => String(ids.worker2));
 
 // SSH connection helper
 function conn(ip: string): command.types.input.remote.ConnectionArgs {
@@ -49,107 +53,181 @@ function conn(ip: string): command.types.input.remote.ConnectionArgs {
 
 // ---------------------------------------------------------------------------
 // Step 1: Start all VMs (fire-and-forget — VMs are owned by proxmox-infra)
+// triggers: vmId — forces re-run when VMs are deleted and recreated
 // ---------------------------------------------------------------------------
 
 const startMaster1 = new command.local.Command("start-master-1", {
   create: pulumi.interpolate`curl -sf -k -X POST -H "Authorization: PVEAPIToken=${pve1ApiToken}" "${pve1Endpoint}/api2/json/nodes/pve/qemu/${master1VmId}/status/start" 2>/dev/null || true`,
+  triggers: [master1VmId],
   interpreter: ["/bin/bash", "-c"],
 });
 const startMaster2 = new command.local.Command("start-master-2", {
   create: pulumi.interpolate`curl -sf -k -X POST -H "Authorization: PVEAPIToken=${pve1ApiToken}" "${pve1Endpoint}/api2/json/nodes/pve/qemu/${master2VmId}/status/start" 2>/dev/null || true`,
+  triggers: [master2VmId],
   interpreter: ["/bin/bash", "-c"],
 });
 const startWorker1 = new command.local.Command("start-worker-1", {
   create: pulumi.interpolate`curl -sf -k -X POST -H "Authorization: PVEAPIToken=${pve1ApiToken}" "${pve1Endpoint}/api2/json/nodes/pve/qemu/${worker1VmId}/status/start" 2>/dev/null || true`,
+  triggers: [worker1VmId],
   interpreter: ["/bin/bash", "-c"],
 });
 const startMaster3 = new command.local.Command("start-master-3", {
   create: pulumi.interpolate`curl -sf -k -X POST -H "Authorization: PVEAPIToken=${pve2ApiToken}" "${pve2Endpoint}/api2/json/nodes/pve-bckp/qemu/${master3VmId}/status/start" 2>/dev/null || true`,
+  triggers: [master3VmId],
   interpreter: ["/bin/bash", "-c"],
 });
 const startWorker2 = new command.local.Command("start-worker-2", {
   create: pulumi.interpolate`curl -sf -k -X POST -H "Authorization: PVEAPIToken=${pve2ApiToken}" "${pve2Endpoint}/api2/json/nodes/pve-bckp/qemu/${worker2VmId}/status/start" 2>/dev/null || true`,
+  triggers: [worker2VmId],
   interpreter: ["/bin/bash", "-c"],
 });
 
-const allStarts = [startMaster1, startMaster2, startMaster3, startWorker1, startWorker2];
+const allStarts = [
+  startMaster1,
+  startMaster2,
+  startMaster3,
+  startWorker1,
+  startWorker2,
+];
 
 // ---------------------------------------------------------------------------
 // Step 2: Wait for SSH on master-1, install k3s
 // ---------------------------------------------------------------------------
 
-const waitMaster1Ssh = new command.local.Command("wait-ssh-master-1", {
-    create: `for i in $(seq 1 60); do nc -z -w 5 ${master1Ip} 22 && exit 0; sleep 5; done; exit 1`,
+const waitMaster1Ssh = new command.local.Command(
+  "wait-ssh-master-1",
+  {
+    create: `for i in $(seq 1 60); do (timeout 5 bash -c "echo > /dev/tcp/${master1Ip}/22") 2>/dev/null && exit 0; sleep 5; done; exit 1`,
+    triggers: [master1VmId],
     interpreter: ["/bin/bash", "-c"],
-}, { dependsOn: allStarts });
+  },
+  { dependsOn: allStarts },
+);
 
-const installMaster1 = new command.remote.Command("install-k3s-master-1", {
+const installMaster1 = new command.remote.Command(
+  "install-k3s-master-1",
+  {
     connection: conn(master1Ip),
-    create: pulumi.interpolate`curl -sfL https://get.k3s.io | K3S_TOKEN='${k3sToken}' sh -s - server --cluster-init --tls-san ${master1Ip}`,
-}, { dependsOn: [waitMaster1Ssh] });
+    create: pulumi.interpolate`curl -sfL https://get.k3s.io | sudo K3S_TOKEN='${k3sToken}' sh -s - server --cluster-init --tls-san ${master1Ip} --node-name k3s-master-1`,
+    triggers: [master1VmId],
+  },
+  { dependsOn: [waitMaster1Ssh] },
+);
 
-const waitK3sMaster1Ready = new command.remote.Command("wait-k3s-master-1-ready", {
+const waitK3sMaster1Ready = new command.remote.Command(
+  "wait-k3s-master-1-ready",
+  {
     connection: conn(master1Ip),
-    create: `for i in $(seq 1 60); do kubectl --kubeconfig /etc/rancher/k3s/k3s.yaml get node k3s-master-1 --no-headers 2>/dev/null | grep -q " Ready" && exit 0; sleep 10; done; exit 1`,
-}, { dependsOn: [installMaster1] });
+    create: `for i in $(seq 1 60); do sudo kubectl --kubeconfig /etc/rancher/k3s/k3s.yaml get node k3s-master-1 --no-headers 2>/dev/null | grep -q " Ready" && exit 0; sleep 10; done; exit 1`,
+    triggers: [master1VmId],
+  },
+  { dependsOn: [installMaster1] },
+);
 
 // ---------------------------------------------------------------------------
 // Step 3: Wait for SSH on remaining masters, join cluster
 // ---------------------------------------------------------------------------
 
-const waitMaster2Ssh = new command.local.Command("wait-ssh-master-2", {
-    create: `for i in $(seq 1 60); do nc -z -w 5 ${master2Ip} 22 && exit 0; sleep 5; done; exit 1`,
+const waitMaster2Ssh = new command.local.Command(
+  "wait-ssh-master-2",
+  {
+    create: `for i in $(seq 1 60); do (timeout 5 bash -c "echo > /dev/tcp/${master2Ip}/22") 2>/dev/null && exit 0; sleep 5; done; exit 1`,
+    triggers: [master2VmId],
     interpreter: ["/bin/bash", "-c"],
-}, { dependsOn: [waitK3sMaster1Ready] });
+  },
+  { dependsOn: [waitK3sMaster1Ready] },
+);
 
-const waitMaster3Ssh = new command.local.Command("wait-ssh-master-3", {
-    create: `for i in $(seq 1 60); do nc -z -w 5 ${master3Ip} 22 && exit 0; sleep 5; done; exit 1`,
+const waitMaster3Ssh = new command.local.Command(
+  "wait-ssh-master-3",
+  {
+    create: `for i in $(seq 1 60); do (timeout 5 bash -c "echo > /dev/tcp/${master3Ip}/22") 2>/dev/null && exit 0; sleep 5; done; exit 1`,
+    triggers: [master3VmId],
     interpreter: ["/bin/bash", "-c"],
-}, { dependsOn: [waitK3sMaster1Ready] });
+  },
+  { dependsOn: [waitK3sMaster1Ready] },
+);
 
-const joinMaster2 = new command.remote.Command("join-k3s-master-2", {
+const joinMaster2 = new command.remote.Command(
+  "join-k3s-master-2",
+  {
     connection: conn(master2Ip),
-    create: pulumi.interpolate`curl -sfL https://get.k3s.io | K3S_TOKEN='${k3sToken}' sh -s - server --server https://${master1Ip}:6443`,
-}, { dependsOn: [waitMaster2Ssh] });
+    create: pulumi.interpolate`curl -sfL https://get.k3s.io | sudo K3S_TOKEN='${k3sToken}' sh -s - server --server https://${master1Ip}:6443 --node-name k3s-master-2`,
+    triggers: [master2VmId],
+  },
+  { dependsOn: [waitMaster2Ssh] },
+);
 
-const joinMaster3 = new command.remote.Command("join-k3s-master-3", {
+const joinMaster3 = new command.remote.Command(
+  "join-k3s-master-3",
+  {
     connection: conn(master3Ip),
-    create: pulumi.interpolate`curl -sfL https://get.k3s.io | K3S_TOKEN='${k3sToken}' sh -s - server --server https://${master1Ip}:6443`,
-}, { dependsOn: [waitMaster3Ssh, joinMaster2] });
+    create: pulumi.interpolate`curl -sfL https://get.k3s.io | sudo K3S_TOKEN='${k3sToken}' sh -s - server --server https://${master1Ip}:6443 --node-name k3s-master-3`,
+    triggers: [master3VmId],
+  },
+  { dependsOn: [waitMaster3Ssh, joinMaster2] },
+);
 
 // ---------------------------------------------------------------------------
 // Step 4: Join worker nodes
 // ---------------------------------------------------------------------------
 
-const waitWorker1Ssh = new command.local.Command("wait-ssh-worker-1", {
-    create: `for i in $(seq 1 60); do nc -z -w 5 ${worker1Ip} 22 && exit 0; sleep 5; done; exit 1`,
+const waitWorker1Ssh = new command.local.Command(
+  "wait-ssh-worker-1",
+  {
+    create: `for i in $(seq 1 60); do (timeout 5 bash -c "echo > /dev/tcp/${worker1Ip}/22") 2>/dev/null && exit 0; sleep 5; done; exit 1`,
+    triggers: [worker1VmId],
     interpreter: ["/bin/bash", "-c"],
-}, { dependsOn: [joinMaster3] });
+  },
+  { dependsOn: [joinMaster3] },
+);
 
-const waitWorker2Ssh = new command.local.Command("wait-ssh-worker-2", {
-    create: `for i in $(seq 1 60); do nc -z -w 5 ${worker2Ip} 22 && exit 0; sleep 5; done; exit 1`,
+const waitWorker2Ssh = new command.local.Command(
+  "wait-ssh-worker-2",
+  {
+    create: `for i in $(seq 1 60); do (timeout 5 bash -c "echo > /dev/tcp/${worker2Ip}/22") 2>/dev/null && exit 0; sleep 5; done; exit 1`,
+    triggers: [worker2VmId],
     interpreter: ["/bin/bash", "-c"],
-}, { dependsOn: [joinMaster3] });
+  },
+  { dependsOn: [joinMaster3] },
+);
 
-const joinWorker1 = new command.remote.Command("join-k3s-worker-1", {
+const joinWorker1 = new command.remote.Command(
+  "join-k3s-worker-1",
+  {
     connection: conn(worker1Ip),
-    create: pulumi.interpolate`curl -sfL https://get.k3s.io | K3S_URL=https://${master1Ip}:6443 K3S_TOKEN='${k3sToken}' sh -s -`,
-}, { dependsOn: [waitWorker1Ssh] });
+    create: pulumi.interpolate`curl -sfL https://get.k3s.io | sudo K3S_URL=https://${master1Ip}:6443 K3S_TOKEN='${k3sToken}' sh -s - --node-name k3s-worker-1`,
+    triggers: [worker1VmId],
+  },
+  { dependsOn: [waitWorker1Ssh] },
+);
 
-const joinWorker2 = new command.remote.Command("join-k3s-worker-2", {
+const joinWorker2 = new command.remote.Command(
+  "join-k3s-worker-2",
+  {
     connection: conn(worker2Ip),
-    create: pulumi.interpolate`curl -sfL https://get.k3s.io | K3S_URL=https://${master1Ip}:6443 K3S_TOKEN='${k3sToken}' sh -s -`,
-}, { dependsOn: [waitWorker2Ssh] });
+    create: pulumi.interpolate`curl -sfL https://get.k3s.io | sudo K3S_URL=https://${master1Ip}:6443 K3S_TOKEN='${k3sToken}' sh -s - --node-name k3s-worker-2`,
+    triggers: [worker2VmId],
+  },
+  { dependsOn: [waitWorker2Ssh] },
+);
 
 // ---------------------------------------------------------------------------
 // Step 5: Read kubeconfig from master-1, patch server URL for external access
 // ---------------------------------------------------------------------------
 
-const getKubeconfig = new command.remote.Command("get-kubeconfig", {
+const getKubeconfig = new command.remote.Command(
+  "get-kubeconfig",
+  {
     connection: conn(master1Ip),
-    create: `cat /etc/rancher/k3s/k3s.yaml`,
-}, { dependsOn: [joinWorker1, joinWorker2] });
+    create: `sudo cat /etc/rancher/k3s/k3s.yaml`,
+    triggers: [master1VmId],
+  },
+  { dependsOn: [joinWorker1, joinWorker2] },
+);
 
 export const kubeconfig = pulumi.secret(
-    getKubeconfig.stdout.apply(kc => kc.replace(/127\.0\.0\.1/g, master1Ip).trim()),
+  getKubeconfig.stdout.apply((kc) =>
+    kc.replace(/127\.0\.0\.1/g, master1Ip).trim(),
+  ),
 );