test(cluster): increase harness timeouts to reduce flaky failures

Cluster bootstrap, join, and convergence waits were capped at 10s,
which is too tight under load (CI, slow VMs). Raise all cluster
harness deadlines to 30s and add a matching nextest slow-timeout
override so the test runner does not terminate the cluster-tests
package before those waits can complete.

Author: farhan-syah

.config/nextest.toml

@@ -43,6 +43,13 @@ threads-required = 'num-test-threads'
 # legitimate startup jitter without hiding real regressions — a
 # genuinely broken test fails twice in a row.
 retries = { backoff = "fixed", count = 2, delay = "1s" }
+# Cluster tests legitimately need more wall-clock than unit tests:
+# a 3-node cluster bring-up plus a multi-node DDL fan-out (e.g.
+# CREATE INDEX backfill across every peer) can stretch past the
+# default 120s ceiling under load without anything being wrong.
+# This override applies ONLY to the cluster-tests package — the
+# default 30s × 4 = 120s ceiling stays in force for unit tests.
+slow-timeout = { period = "30s", terminate-after = 6 }
 
 [test-groups]
 cluster = { max-threads = 1 }

README.md

@@ -210,11 +210,11 @@ cargo nextest run --all-features
 
 ## Release Status
 
-NodeDB Origin is in **public beta** as of **v0.1.0 (2026-05-01)**. All eight engines are feature-complete and covered by tests. The wire protocols (pgwire, HTTP, native MessagePack, RESP, ILP, WebSocket) are stable — clients written against 0.1.0 will keep working through 1.0.
+NodeDB Origin is in **public beta** as of **v0.1.0 (2026-05-07)**. All eight engines are feature-complete and covered by tests. The wire protocols (pgwire, HTTP, native MessagePack, RESP, ILP, WebSocket) are stable — clients written against 0.1.0 will keep working through 1.0.
 
 **v0.1.0 — Beta (today).** Build new products on it. The public surface (SQL dialect, wire protocols, configuration) is stable; expect internal changes (storage layout, on-disk format, replication internals) between minor releases. Patch and minor bumps will land as needed — if something requires a 0.2, we ship a 0.2. The two months between beta and 1.0 are deliberately for real workloads to surface edge cases we can't manufacture in-house.
 
-**v1.0.0 — Production-ready (target: 2026-07-01).** What 1.0 guarantees:
+**v1.0.0 — Production-ready (target: 2026-07-07).** What 1.0 guarantees:
 
 - **API & SQL stability** — semver from 1.0 onward. No breaking SQL or client-API changes within a major.
 - **Wire protocol stability** — pgwire, HTTP, native MessagePack, RESP, ILP, WebSocket frozen.

nodedb-test-support/src/cluster_harness/cluster.rs

@@ -52,10 +52,10 @@ impl TestCluster {
         // short under heavy host load (e.g. 500+ parallel unit tests
         // sharing the same CPU pool), causing peers to dial before
         // node 1's transport was ready — failing topology convergence.
-        let deadline = std::time::Instant::now() + Duration::from_secs(10);
+        let deadline = std::time::Instant::now() + Duration::from_secs(30);
         while node1.topology_size() < 1 {
             if std::time::Instant::now() >= deadline {
-                return Err("node 1 failed to bootstrap within 10s".into());
+                return Err("node 1 failed to bootstrap within 30s".into());
             }
             tokio::time::sleep(Duration::from_millis(20)).await;
         }
@@ -67,10 +67,10 @@ impl TestCluster {
         // Under load, spawning both peers simultaneously can overwhelm the
         // bootstrap leader's join handler, causing neither join to complete
         // within the topology convergence deadline.
-        let deadline = std::time::Instant::now() + Duration::from_secs(10);
+        let deadline = std::time::Instant::now() + Duration::from_secs(30);
         while node1.topology_size() < 2 {
             if std::time::Instant::now() >= deadline {
-                return Err("node 2 failed to join within 10s".into());
+                return Err("node 2 failed to join within 30s".into());
             }
             tokio::time::sleep(Duration::from_millis(20)).await;
         }
@@ -83,7 +83,7 @@ impl TestCluster {
 
         wait_for(
             "all 3 nodes report topology_size == 3",
-            Duration::from_secs(10),
+            Duration::from_secs(30),
             Duration::from_millis(50),
             || cluster.nodes.iter().all(|n| n.topology_size() == 3),
         )
@@ -114,7 +114,7 @@ impl TestCluster {
         // replication.
         wait_for(
             "all 3 nodes exit rolling-upgrade compat mode",
-            Duration::from_secs(10),
+            Duration::from_secs(30),
             Duration::from_millis(20),
             || {
                 cluster.nodes.iter().all(|n| {
@@ -145,7 +145,7 @@ impl TestCluster {
         // wasted CI minutes on cleanup of a doomed cluster bringup.
         wait_for(
             "metadata group has stable leader visible on every node",
-            Duration::from_secs(10),
+            Duration::from_secs(30),
             Duration::from_millis(20),
             || {
                 let leaders: Vec<u64> = cluster
@@ -182,7 +182,7 @@ impl TestCluster {
         // window deterministically.
         wait_for(
             "every Raft group has a stable leader visible on every node",
-            Duration::from_secs(10),
+            Duration::from_secs(30),
             Duration::from_millis(20),
             || {
                 // Snapshot every node's per-group leader view. A group