Files
lab/bastion/tests/integration/helpers/libvirt.ts

220 lines
6.8 KiB
TypeScript
Raw Normal View History

feat: install logging, error trapping, PXE/ISO integration tests Kickstart installs on real hardware failed silently — no error reporting, only 3 progress callbacks, zero log streaming. This overhaul makes every install fully observable. Kickstart improvements: - Error trapping in %pre and %post (trap ERR sends failure details to bastion) - 12+ granular progress stages (was 3): SSH, hostname, k3s prep, EFI boot, metadata - Background log streamer: tails %post output and batch-sends to /api/log - bastion_log() function for explicit log lines from kickstart scripts Bastion API: - POST /api/log — receives raw log lines from kickstart (single or batch) - InstallLogBuffer — per-MAC ring buffer (2000 lines) + file persistence - GET /api/logs/:mac — now returns log_lines + log_total alongside stages - SSE /api/logs/:mac/follow — uses named events (event: stage vs event: log) - Progress events forwarded to labd via bastion-progress WebSocket message - Post-provision k3s logs routed through progressBus (was console-only) dnsmasq fixes found during VM testing: - HTTP Boot filename: ipxe-real.efi → ipxe.efi (leftover from old 2-stage approach) - pxe-service directives: only in proxy mode (breaks OVMF PXE in full mode) - PXEClient vendor class echo for UEFI firmware compatibility Integration tests: - PXE boot test: blank UEFI VM → dnsmasq → HTTP Boot → iPXE → bastion → install - ISO boot test: blank VM boots from bastion-generated ISO → same flow - Shared helpers: pxe-network (no DHCP, nftables fix), pxe-vm (UEFI + ISO boot) - test-provision.sh: runs both PXE + ISO tests with prerequisite checks - 250GB sparse QCOW2 disk (LVM layout needs ~204GB) 201 unit tests passing (11 new). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-26 22:26:33 +00:00
// Libvirt VM lifecycle management for integration tests.
import { execSync, spawnSync, type SpawnSyncReturns } from "node:child_process";
import { existsSync, mkdirSync, writeFileSync, unlinkSync } from "node:fs";
import { join } from "node:path";
const IMAGE_DIR = "/var/lib/libvirt/images";
// Cloud-init ISOs must be in a path accessible to the host's libvirtd,
// so we use the shared images directory (not /tmp which may be container-only).
const CLOUD_INIT_DIR = "/var/lib/libvirt/images/lab-cloud-init";
// When running as root (inside container or via sudo), don't prefix with sudo.
const IS_ROOT = process.getuid?.() === 0;
/** Run a shell command, prefixing with sudo if not root. */
function run(cmd: string, opts?: { timeout?: number }): string {
const full = IS_ROOT ? cmd : `sudo ${cmd}`;
return execSync(full, { encoding: "utf-8", stdio: "pipe", timeout: opts?.timeout ?? 60_000 });
}
/** Spawn a command, prefixing args with sudo if not root. */
function virsh(...args: string[]): SpawnSyncReturns<string> {
const cmd = IS_ROOT ? "virsh" : "sudo";
const finalArgs = IS_ROOT ? args : ["virsh", ...args];
return spawnSync(cmd, finalArgs, { encoding: "utf-8", stdio: "pipe", timeout: 30_000 });
}
export interface VmConfig {
name: string;
memory: number; // MB
vcpus: number;
diskSize: number; // GB
network: string; // libvirt network name
cloudImageUrl: string;
sshPubKey: string; // content of authorized key
userData?: string; // custom cloud-init user-data
}
export function log(msg: string): void {
const ts = new Date().toISOString().slice(11, 19);
console.log(` [${ts}] ${msg}`);
}
/** Download a cloud image if not already cached. */
export function ensureCloudImage(url: string, name: string): string {
const dest = join(IMAGE_DIR, `${name}.qcow2`);
if (existsSync(dest)) {
log(`Cloud image cached: ${dest}`);
return dest;
}
log(`Downloading cloud image: ${url}`);
run(`curl -L -f -o "${dest}" "${url}"`, { timeout: 300_000 });
return dest;
}
/** Create a cloud-init ISO for a VM. */
export function createCloudInitIso(vmName: string, config: VmConfig): string {
mkdirSync(CLOUD_INIT_DIR, { recursive: true });
const dir = join(CLOUD_INIT_DIR, vmName);
mkdirSync(dir, { recursive: true });
const userData = config.userData ?? `#cloud-config
hostname: ${vmName}
manage_etc_hosts: true
users:
- default
- name: fedora
sudo: ALL=(ALL) NOPASSWD:ALL
shell: /bin/bash
ssh_authorized_keys:
- ${config.sshPubKey}
ssh_pwauth: false
package_update: false
packages:
- curl
- socat
- conntrack-tools
- ethtool
- iptables-nft
runcmd:
- modprobe br_netfilter
- modprobe overlay
- |
cat > /etc/sysctl.d/90-k3s.conf << 'EOF'
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward = 1
vm.panic_on_oom = 0
vm.overcommit_memory = 1
kernel.panic = 10
kernel.panic_on_oops = 1
EOF
- sysctl --system
- swapoff -a
- systemctl disable --now firewalld 2>/dev/null || true
- systemctl disable --now ufw 2>/dev/null || true
`;
const metaData = `instance-id: ${vmName}\nlocal-hostname: ${vmName}\n`;
writeFileSync(join(dir, "user-data"), userData);
writeFileSync(join(dir, "meta-data"), metaData);
const isoPath = join(CLOUD_INIT_DIR, `${vmName}-cloud-init.iso`);
execSync(
`genisoimage -output "${isoPath}" -volid cidata -joliet -rock "${dir}/user-data" "${dir}/meta-data" 2>/dev/null || ` +
`mkisofs -output "${isoPath}" -volid cidata -joliet -rock "${dir}/user-data" "${dir}/meta-data" 2>/dev/null || ` +
`xorrisofs -output "${isoPath}" -volid cidata -joliet -rock "${dir}/user-data" "${dir}/meta-data"`,
{ stdio: "pipe" },
);
return isoPath;
}
/** Create and start a VM from a cloud image with cloud-init. */
export function createVm(config: VmConfig): void {
destroyVm(config.name);
log(`Creating VM: ${config.name} (${config.memory}MB RAM, ${config.vcpus} vCPU, ${config.diskSize}GB disk)`);
const baseImage = ensureCloudImage(config.cloudImageUrl, `${config.name}-base`);
const diskPath = join(IMAGE_DIR, `${config.name}.qcow2`);
run(`cp "${baseImage}" "${diskPath}"`);
run(`qemu-img resize "${diskPath}" ${config.diskSize}G`);
const cloudInitIso = createCloudInitIso(config.name, config);
const virtInstallArgs = [
"virt-install",
`--name=${config.name}`,
`--memory=${config.memory}`,
`--vcpus=${config.vcpus}`,
`--disk=path=${diskPath},format=qcow2`,
`--disk=path=${cloudInitIso},device=cdrom`,
`--network=network=${config.network},model=virtio`,
"--os-variant=generic",
"--import",
"--noautoconsole",
"--wait=0",
];
log(`Running: virt-install --name=${config.name} ...`);
run(virtInstallArgs.join(" "));
log(`VM ${config.name} created and starting`);
}
/** Destroy a VM and remove its storage. */
export function destroyVm(name: string): void {
const result = virsh("dominfo", name);
if (result.status !== 0) return;
log(`Destroying VM: ${name}`);
virsh("destroy", name);
virsh("undefine", name, "--remove-all-storage");
const isoPath = join(CLOUD_INIT_DIR, `${name}-cloud-init.iso`);
try { unlinkSync(isoPath); } catch { /* ignore */ }
}
/** Get the IP address of a running VM. */
export function getVmIp(name: string): string | null {
try {
// Try with agent first, then without
let output = virsh("domifaddr", name, "--source", "agent").stdout;
if (!output || !output.includes(".")) {
output = virsh("domifaddr", name).stdout;
}
const match = output.match(/(\d+\.\d+\.\d+\.\d+)/);
return match ? match[1] : null;
} catch {
return null;
}
}
/** Wait for a VM to get an IP address. */
export async function waitForVmIp(name: string, timeoutMs: number): Promise<string> {
const start = Date.now();
while (Date.now() - start < timeoutMs) {
const ip = getVmIp(name);
if (ip) {
log(`VM ${name} got IP: ${ip}`);
return ip;
}
await sleep(2000);
}
throw new Error(`VM ${name} did not get an IP within ${timeoutMs}ms`);
}
/** Wait for SSH to become available on a host. */
export async function waitForSsh(
ip: string,
user: string,
timeoutMs: number,
keyPath?: string,
): Promise<void> {
const start = Date.now();
while (Date.now() - start < timeoutMs) {
const result = spawnSync("ssh", [
"-o", "StrictHostKeyChecking=no",
"-o", "ConnectTimeout=3",
"-o", "BatchMode=yes",
...(keyPath ? ["-i", keyPath] : []),
`${user}@${ip}`,
"echo ok",
], { encoding: "utf-8", stdio: "pipe", timeout: 10_000 });
if (result.status === 0 && result.stdout.includes("ok")) {
log(`SSH ready on ${ip}`);
return;
}
await sleep(3000);
}
throw new Error(`SSH not available on ${ip} within ${timeoutMs}ms`);
}
function sleep(ms: number): Promise<void> {
return new Promise((r) => setTimeout(r, ms));
}