src/lifecycle/egress.ts

@@ -131,19 +131,28 @@ async function reaches(name: string, waitSeconds: number): Promise<string | null
  * explicitly and nothing else defaults.
  */
 /**
- * Keep the uplink as the resolver and give it company.
+ * Give the machine's uplink more than one resolver, through the thing that owns resolvers.
  *
- * Appended rather than replacing: the uplink is still asked first and still proves the modelled
+ * **Not by writing /etc/resolv.conf**, which was the first attempt and was wrong: on these images
- * path. The fallbacks only answer when it does not, which under a four-machine image pull is a
+ * that path is a symlink managed by systemd-resolved, so a file written over it is either reverted
- * thing that happens and has ended three runs.
+ * or breaks the link. Checked on a running machine rather than assumed.
+ *
+ * The shape of the problem is visible in `resolvectl status`: the machine has sensible global
+ * fallbacks, and the *link* carrying the default route has exactly one server — the uplink gateway.
+ * resolved will not reach for a global fallback while the link it is using has a server of its own,
+ * so one unanswered packet is one failed lookup. Under four machines pulling images at once that
+ * happens, and it has ended three runs long after the egress check passed.
+ *
+ * The uplink stays first, so the modelled path is still the one used and still proven by the check
+ * above. The others answer only when it does not.
  */
 async function resilientResolver(name: string): Promise<void> {
   await incus([
     "exec", name, "--", "sh", "-c",
-    `via=$(ip -4 route show default | awk '{print $3}' | head -n1); ` +
+    `link=$(ip -4 route show default | awk '{print $5}' | head -n1); ` +
-      `{ [ -n "$via" ] && printf 'nameserver %s\\n' "$via"; ` +
+      `via=$(ip -4 route show default | awk '{print $3}' | head -n1); ` +
-      `printf 'nameserver 1.1.1.1\\nnameserver 8.8.8.8\\n'; ` +
+      `if [ -n "$link" ] && command -v resolvectl >/dev/null 2>&1; then ` +
-      `printf 'options timeout:2 attempts:3\\n'; } > /etc/resolv.conf; true`,
+      `resolvectl dns "$link" $via 1.1.1.1 8.8.8.8 >/dev/null 2>&1 || true; fi; true`,
   ], 30_000);
 }