Task manager: release task locks on stop, recheck claims held elsewhere

A cluster rehearsal (PostgreSQL + NATS) left index tasks pending well
past the one-hour task lock after the node that claimed them was stopped
or killed. The exact cause there isn't confirmed; this closes every path
found in the task manager that stretches a takeover past the lock, or
keeps a task claimed without running it:

- A graceful stop never released the locks it held, so every task the
  node had claimed stayed blocked for an hour. The server now tracks the
  locks it holds (common::ipc::TaskLocks) and, once the shutdown signal
  arrives, stops claiming and releases them before exiting.
- A node that failed to claim a task (another node held it) set its own
  local hold for a full lock lifetime from that scan. If the holder
  claimed it just after the scan began, or ran on a clock ahead, that
  hold ran out a moment before the lock did and was set for another
  hour: two hours in all. Such claims are now tried again every five
  minutes (a twelfth of the lock lifetime), and the task manager wakes
  up for them: before, a node without a coordinator could sleep up to
  five minutes past the recheck, or until something else woke it.
- A worker that panicked took its task type down on that node for good,
  while the scan kept claiming that type's tasks and failing to hand them
  over, re-taking each lock as it expired and so starving every other
  node of them. Each batch now runs on a task of its own; a panic is
  logged, the batch's locks are released and the worker carries on. A
  failed hand-over releases the lock too.
- A claimed task the worker couldn't read, or found gone, kept its lock
  for the hour. It is released.
- An IndexDocument task for a file (not indexed) returned no result,
  which shifted every later result in the batch onto the wrong task in
  update_tasks. It returns Ignored. Nothing queues such a task today.

The lock lifetime stays one hour; it now lives per server so the tests
can shorten it.

store::task_locks::task_lock_tests plays a second node by writing its
locks straight into the in-memory store: tasks it claimed and abandoned
run here once its locks expire, including locks that outlive this node's
view of them, and a graceful stop hands this node's locks back at once
and claims nothing more. It passes on RocksDB, SQLite and PostgreSQL.
With the old recheck it fails.

tests/src/store/mod.rs

@@ -21,6 +21,7 @@ pub mod replica_cluster; // inbuxa: read replicas across nodes
 pub mod scaleout; // inbuxa: scale-out storage
 #[cfg(any(feature = "postgres", feature = "mysql"))]
 pub mod sql_timeout;
+pub mod task_locks; // inbuxa: task locks across nodes
 
 use crate::utils::server::TestServerBuilder;
 use std::io::Read;

tests/src/store/task_locks.rs

@@ -0,0 +1,184 @@
+/*
+ * SPDX-FileCopyrightText: 2026 Coffey Labs
+ *
+ * SPDX-License-Identifier: AGPL-3.0-only
+ */
+
+//! Task locks across nodes: tasks claimed by a node that then disappears
+//! run elsewhere once its locks expire, and a node that stops gracefully
+//! hands its locks back at once. The other node is played by writing its
+//! locks straight into the shared in-memory store, as a node that claimed
+//! the tasks and died leaves them.
+
+use crate::utils::server::TestServerBuilder;
+use common::{KV_LOCK_TASK, Server};
+use registry::schema::{
+    enums::IndexDocumentType,
+    structs::{Task, TaskIndexDocument, TaskStatus},
+};
+use services::task_manager::lock::{TaskLockManager, release_task_locks};
+use std::time::{Duration, Instant};
+use store::{
+    ValueKey,
+    write::{BatchBuilder, TaskQueueClass, ValueClass},
+};
+use utils::snowflake::SnowflakeIdGenerator;
+
+// Short enough for a test, long enough that the recheck interval (a twelfth
+// of it) is well below it
+const LOCK_EXPIRY: u64 = 12;
+
+#[tokio::test(flavor = "multi_thread")]
+pub async fn task_lock_tests() {
+    let test = TestServerBuilder::new("task_lock_tests")
+        .await
+        .build()
+        .await;
+    let server = test.server.clone();
+    println!(
+        "Running task lock tests on {}...",
+        std::env::var("STORE").unwrap_or_default()
+    );
+    server.inner.ipc.task_locks.set_expiry(LOCK_EXPIRY);
+
+    // 1. Another node claimed the tasks and died. Its locks block them until
+    // they expire; then this node runs them, without waiting for anything
+    // else to wake it
+    let ids = new_task_ids(4);
+    for id in &ids {
+        assert!(foreign_lock(&server, *id, LOCK_EXPIRY).await);
+    }
+    schedule(&server, &ids).await;
+    let started = Instant::now();
+    server.notify_task_queue();
+    tokio::time::sleep(Duration::from_secs(3)).await;
+    assert_eq!(pending(&server, &ids).await, ids.len(), "held by the other node");
+    wait_until_done(&server, &ids, Duration::from_secs(LOCK_EXPIRY + 10)).await;
+    let elapsed = started.elapsed();
+    assert!(
+        elapsed >= Duration::from_secs(LOCK_EXPIRY - 2),
+        "ran before the other node's locks expired: {elapsed:?}"
+    );
+
+    // 2. The other node's locks outlive what this node expects: claimed just
+    // after this node looked, or by a node whose clock runs ahead. This node
+    // keeps checking at the recheck interval, so the tasks run soon after
+    // those locks expire, not a whole lock lifetime later
+    let held_for = LOCK_EXPIRY + LOCK_EXPIRY / 2;
+    let ids = new_task_ids(4);
+    for id in &ids {
+        assert!(foreign_lock(&server, *id, held_for).await);
+    }
+    schedule(&server, &ids).await;
+    let started = Instant::now();
+    server.notify_task_queue();
+    wait_until_done(&server, &ids, Duration::from_secs(held_for + 8)).await;
+    let elapsed = started.elapsed();
+    assert!(
+        elapsed >= Duration::from_secs(held_for - 2),
+        "ran before the other node's locks expired: {elapsed:?}"
+    );
+
+    // 3. A graceful stop releases the locks this node holds: another node
+    // can claim those tasks at once, and this one claims nothing more
+    let ids = new_task_ids(3);
+    for id in &ids {
+        assert!(server.try_lock_task(*id).await, "claim {id}");
+    }
+    assert_eq!(server.inner.ipc.task_locks.held(), ids.len());
+    for id in &ids {
+        assert!(
+            !foreign_lock(&server, *id, LOCK_EXPIRY).await,
+            "held while this node runs"
+        );
+    }
+    assert_eq!(release_task_locks(&server).await, ids.len());
+    assert_eq!(server.inner.ipc.task_locks.held(), 0);
+    for id in &ids {
+        assert!(
+            foreign_lock(&server, *id, LOCK_EXPIRY).await,
+            "released on stop: {id}"
+        );
+    }
+    let [id] = new_task_ids(1)[..] else {
+        unreachable!()
+    };
+    assert!(!server.try_lock_task(id).await, "a stopping node claims nothing");
+
+    for id in ids {
+        let _ = server
+            .in_memory_store()
+            .remove_lock(KV_LOCK_TASK, &id.to_be_bytes())
+            .await;
+    }
+    if test.is_reset() {
+        test.temp_dir.delete();
+    }
+}
+
+fn new_task_ids(count: usize) -> Vec<u64> {
+    (0..count)
+        .map(|_| SnowflakeIdGenerator::global_id().unwrap())
+        .collect()
+}
+
+/// The other node's claim on a task, as its task manager takes it.
+async fn foreign_lock(server: &Server, id: u64, seconds: u64) -> bool {
+    server
+        .in_memory_store()
+        .try_lock(KV_LOCK_TASK, &id.to_be_bytes(), seconds)
+        .await
+        .unwrap()
+}
+
+/// Unindex tasks for files that don't exist: files aren't search-indexed and
+/// there is no undelete note, so running one only drops it from the queue.
+async fn schedule(server: &Server, ids: &[u64]) {
+    let mut batch = BatchBuilder::new();
+    for (n, id) in ids.iter().enumerate() {
+        batch.schedule_task_with_id(
+            *id,
+            Task::UnindexDocument(TaskIndexDocument {
+                account_id: 0u32.into(),
+                document_id: (u32::MAX - n as u32).into(),
+                document_type: IndexDocumentType::File,
+                status: TaskStatus::now(),
+            }),
+        );
+    }
+    server.store().write(batch.build_all()).await.unwrap();
+}
+
+async fn pending(server: &Server, ids: &[u64]) -> usize {
+    let mut count = 0;
+    for id in ids {
+        if server
+            .store()
+            .get_value::<Task>(ValueKey::from(ValueClass::TaskQueue(
+                TaskQueueClass::Task { id: *id },
+            )))
+            .await
+            .unwrap()
+            .is_some()
+        {
+            count += 1;
+        }
+    }
+    count
+}
+
+async fn wait_until_done(server: &Server, ids: &[u64], within: Duration) {
+    let started = Instant::now();
+    loop {
+        let left = pending(server, ids).await;
+        if left == 0 {
+            return;
+        }
+        assert!(
+            started.elapsed() < within,
+            "{left} task(s) still pending after {:?}",
+            started.elapsed()
+        );
+        tokio::time::sleep(Duration::from_millis(250)).await;
+    }
+}