Fix: Diagnostic could be improved

Solution:
- Add checks on the host connectivity, to help with nodes without IPv6 connectivity
- Add links to troubleshooting docs
- Move IPv6 Egress in details section, it was confusing to have it on the main page.

src/aleph/vm/conf.py

@@ -127,6 +127,11 @@ class Settings(BaseSettings):
     WATCH_FOR_UPDATES = True
 
     API_SERVER = "https://official.aleph.cloud"
+    # Connect to the Quad9 VPN provider using their IPv4 and IPv6 addresses.
+    CONNECTIVITY_IPV4_URL = "https://9.9.9.9/"
+    CONNECTIVITY_IPV6_URL = "https://[2620:fe::fe]/"
+    CONNECTIVITY_DNS_HOSTNAME = "example.org"
+
     USE_JAILER = True
     # System logs make boot ~2x slower
     PRINT_SYSTEM_LOGS = False

src/aleph/vm/orchestrator/supervisor.py

@@ -28,6 +28,7 @@
     run_code_from_hostname,
     run_code_from_path,
     status_check_fastapi,
+    status_check_host,
     status_check_version,
     status_public_config,
     update_allocations,
@@ -90,6 +91,7 @@ async def allow_cors_on_endpoint(request: web.Request):
             allow_cors_on_endpoint,
         ),
         web.get("/status/check/fastapi", status_check_fastapi),
+        web.get("/status/check/host", status_check_host),
         web.get("/status/check/version", status_check_version),
         web.get("/status/config", status_public_config),
         web.static("/static", Path(__file__).parent / "views/static"),

src/aleph/vm/orchestrator/views/__init__.py

@@ -26,6 +26,14 @@
 from aleph.vm.orchestrator.pubsub import PubSub
 from aleph.vm.orchestrator.resources import Allocation
 from aleph.vm.orchestrator.run import run_code_on_request, start_persistent_vm
+from aleph.vm.orchestrator.views.host_status import (
+    check_dns_ipv4,
+    check_dns_ipv6,
+    check_domain_resolution_ipv4,
+    check_domain_resolution_ipv6,
+    check_host_egress_ipv4,
+    check_host_egress_ipv6,
+)
 from aleph.vm.pool import VmPool
 from aleph.vm.utils import (
     HostNotFoundError,
@@ -169,6 +177,25 @@ async def status_check_fastapi(request: web.Request):
         return web.json_response(result, status=200 if all(result.values()) else 503)
 
 
+async def status_check_host(request: web.Request):
+    """Check that the platform is supported and configured correctly"""
+
+    result = {
+        "ipv4": {
+            "egress": await check_host_egress_ipv4(),
+            "dns": await check_dns_ipv4(),
+            "domain": await check_domain_resolution_ipv4(),
+        },
+        "ipv6": {
+            "egress": await check_host_egress_ipv6(),
+            "dns": await check_dns_ipv6(),
+            "domain": await check_domain_resolution_ipv6(),
+        },
+    }
+    result_status = 200 if all(result["ipv4"].values()) and all(result["ipv6"].values()) else 503
+    return web.json_response(result, status=result_status)
+
+
 async def status_check_version(request: web.Request):
     """Check if the software is running a version equal or newer than the given one"""
     reference_str: Optional[str] = request.query.get("reference")

src/aleph/vm/orchestrator/views/host_status.py

@@ -0,0 +1,91 @@
+import socket
+from typing import Literal, Union
+
+import aiohttp
+
+from aleph.vm.conf import settings
+
+
+async def check_ip_connectivity(url: str, socket_family: socket.AddressFamily = socket.AF_UNSPEC) -> bool:
+    timeout = aiohttp.ClientTimeout(total=5)
+    async with aiohttp.ClientSession(connector=aiohttp.TCPConnector(), timeout=timeout) as session:
+        async with session.get(url) as resp:
+            # We expect the Quad9 endpoints to return a 404 error, but other endpoints may return a 200
+            if resp.status not in (200, 404):
+                resp.raise_for_status()
+            return True
+
+
+async def check_host_egress_ipv4() -> bool:
+    """Check if the host has IPv4 connectivity."""
+    try:
+        return await check_ip_connectivity(settings.CONNECTIVITY_IPV4_URL)
+    except TimeoutError as exc:
+        print(f"IPv4 connectivity test failed: {exc}")
+        return False
+
+
+async def check_host_egress_ipv6() -> bool:
+    """Check if the host has IPv6 connectivity."""
+    try:
+        return await check_ip_connectivity(settings.CONNECTIVITY_IPV6_URL)
+    except TimeoutError as exc:
+        print(f"IPv6 connectivity test failed: {exc}")
+        return False
+
+
+async def resolve_dns(hostname: str) -> dict:
+    """Resolve a hostname to an IP address."""
+    info_inet, info_inet6 = socket.getaddrinfo(hostname, 80, proto=socket.IPPROTO_TCP)
+    ipv4 = info_inet[4][0]
+    ipv6 = info_inet6[4][0]
+    return {
+        "ipv4": ipv4,
+        "ipv6": ipv6,
+    }
+
+
+async def check_dns_ipv4() -> bool:
+    """Check if DNS resolution is working via IPv4."""
+    resolution = await resolve_dns(settings.CONNECTIVITY_DNS_HOSTNAME)
+    ipv4 = resolution["ipv4"]
+    return bool(ipv4)
+
+
+async def check_dns_ipv6() -> bool:
+    """Check if DNS resolution is working via IPv6."""
+    resolution = await resolve_dns(settings.CONNECTIVITY_DNS_HOSTNAME)
+    ipv6 = resolution["ipv6"]
+    return bool(ipv6)
+
+
+async def check_domain_resolution_ipv4() -> bool:
+    """Check if the host's hostname resolves to an IPv4 address."""
+    resolution = await resolve_dns(settings.DOMAIN_NAME)
+    ipv4 = resolution["ipv4"]
+    return bool(ipv4)
+
+
+async def check_domain_resolution_ipv6() -> bool:
+    """Check if the host's hostname resolves to an IPv6 address."""
+    resolution = await resolve_dns(settings.DOMAIN_NAME)
+    ipv6 = resolution["ipv6"]
+    return False
+
+
+async def check_domain_ipv4() -> bool:
+    """Check if the host's hostname is accessible via IPv4."""
+    try:
+        return await check_ip_connectivity(settings.DOMAIN_NAME, socket.AF_INET)
+    except TimeoutError as exc:
+        print(f"IPv4 connectivity test failed: {exc}")
+        return False
+
+
+async def check_domain_ipv6() -> bool:
+    """Check if the host's hostname is accessible via IPv6."""
+    try:
+        return await check_ip_connectivity(settings.DOMAIN_NAME, socket.AF_INET6)
+    except TimeoutError as exc:
+        print(f"IPv6 connectivity test failed: {exc}")
+        return False

src/aleph/vm/orchestrator/views/static/helpers.js

@@ -1,4 +1,4 @@
-async function fetchApiStatus () {
+async function fetchFastapiCheckStatus () {
     const q = await fetch('/status/check/fastapi');
     let res = {
         status: q.status,
@@ -12,8 +12,10 @@ async function fetchApiStatus () {
             case 503:
                 res.status = " is not working properly ❌";
                 res.details = await q.json();
+                break;
             case 500:
                 res.status = " ❌ Failed";
+                break;
             default:
                 res.status = q.status;
         }
@@ -22,6 +24,36 @@ async function fetchApiStatus () {
     return res;
 }
 
+async function fetchHostCheckStatus () {
+    const q = await fetch('/status/check/host');
+    let res = {
+        status: q.status,
+        details: []
+    }
+    if(q.ok){
+        res.status = " is working properly ✅";
+    }
+    else {
+        switch(Number(q.status)){
+            case 503:
+                res.status = " is not working properly ❌";
+                res.details = await q.json();
+                break;
+            case 500:
+                res.status = " ❌ Failed";
+                break;
+            default:
+                res.status = q.status;
+        }
+    }
+
+    return res;
+}
+
+function objectToString (obj) {
+    return Object.entries(obj).reduce((acc, [k, v]) => acc + `<li>${k}: <span style="color: ${v ? 'green' : 'red'}">${v}</span></li>\n`, '');
+}
+
 const buildQueryParams = (params) => Object.entries(params).reduce((acc, [k, v]) => acc + `${k}=${v}&`, '?').slice(0, -1);
 
 const isLatestRelease = async () => {

src/aleph/vm/orchestrator/views/static/main.css

@@ -1,11 +1,14 @@
 body {
     font-family: IBM Plex Regular, monospace;
+    white-space: normal;
     margin: auto;
+    max-width: 800px;
 }
 
 main {
     width: 90vw;
     margin: 2vh auto;
+    max-width: 800px;
 }
 
 progress {
@@ -36,29 +39,29 @@ progress {
 
 @keyframes move {
     0% {
-        height: 20px;
+        height: 10px;
     }
 
     50% {
-        height: 10px;
+        height: 5px;
     }
 
     100% {
-        height: 20px;
+        height: 10px;
     }
 }
 
 @keyframes move2 {
     0% {
-        height: 10px;
+        height: 5px;
     }
 
     50% {
-        height: 20px;
+        height: 10px;
     }
 
     100% {
-        height: 10px;
+        height: 5px;
     }
 }
 
@@ -97,4 +100,4 @@ progress {
 footer{
     font-size: 70%;
     opacity: .75;
-}
+}