phacility/phabricator · warning · DiffusionDaemonLockException

Another process is currently updating repository "%s", skipp

Error message

Another process is currently updating repository "%s", skipping pull.

What it means

Thrown by PhabricatorRepositoryPullEngine::pullRepository() when the per-repository repo.pull lock cannot be acquired because another process is already pulling this repository. Pulls are serialized per repository so two processes never write the same working copy concurrently; on contention the engine throws DiffusionDaemonLockException so the caller (pull daemon or task worker) skips this round and retries later.

Source

Thrown at src/applications/repository/engine/PhabricatorRepositoryPullEngine.php:32

 * @task svn      Pulling Subversion Working Copies
 * @task internal Internals
 */
final class PhabricatorRepositoryPullEngine
  extends PhabricatorRepositoryEngine {


/* -(  Pulling Working Copies  )--------------------------------------------- */


  public function pullRepository() {
    $repository = $this->getRepository();

    $lock = $this->newRepositoryLock($repository, 'repo.pull', true);

    try {
      $lock->lock();
    } catch (PhutilLockException $ex) {
      throw new DiffusionDaemonLockException(
        pht(
          'Another process is currently updating repository "%s", '.
          'skipping pull.',
          $repository->getDisplayName()));
    }

    try {
      $result = $this->pullRepositoryWithLock();
    } catch (Exception $ex) {
      $lock->unlock();
      throw $ex;
    }

    $lock->unlock();

    return $result;
  }

View on GitHub (pinned to 5720a38cfe)

Solutions

Treat as transient in daemon logs; the next pull cycle retries automatically
Check 'phd status' and running processes for a pull of the same repository that is still active or hung
If a pull is hung (large repo, slow network), let it finish or terminate it cleanly, then retry
Restart daemons with 'phd restart' if a stale lock is suspected

Defensive patterns

Strategy: retry

Try / catch

try {
  id(new PhabricatorRepositoryPullEngine())
    ->setRepository($repository)
    ->pullRepository();
} catch (DiffusionDaemonLockException $ex) {
  // repo.pull held elsewhere; skip and let the next cycle retry
  return;
}

Prevention

Avoid manual 'bin/repository pull' while pull daemons run
For very large repos, expect long pull windows and tolerate lock skips
Use 'phd restart' after killing workers to clear stale pull locks

When it happens

Trigger: A pull task for the repository still in flight when another task for the same repository starts; manual 'bin/repository pull' racing the daemon; slow pulls on large repos or high-latency networks widening the overlap window; stale lock from a killed pull process.

Common situations: Very large repositories whose pulls outlast the scheduling interval; administrators manually pulling to debug while daemons run; lingering worker processes after a phd restart.

Related errors

AI-assisted analysis of phacility/phabricator@5720a38cfe (2026-08-21). Data as JSON: /api/errors/1c40d959928cccf1. Report an issue: GitHub.