Skip to content

fix: retry transient InnoDB deadlocks during option-store lock acquisition #580

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
chubes4 merged 1 commit into from
Sep 29, 2026
Merged

fix: retry transient InnoDB deadlocks during option-store lock acquisition #580

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 19 additions & 2 deletions src/Runtime/class-wp-agent-option-run-control-store.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -281,7 +281,11 @@ private function acquire_lock( \wpdb $db, string $table, string $lock_key ): str
return $lock_value;
}
if ( false === $created ) {
throw new WP_Agent_Run_Control_Store_Exception( 'Atomic run-control lock creation failed.' );
if ( ! $this->is_transient_lock_contention( $db ) ) {
throw new WP_Agent_Run_Control_Store_Exception( 'Atomic run-control lock creation failed.' );
}
usleep( 50000 );
continue;
}

$current = $this->get_var( $db, $db->prepare( 'SELECT option_value FROM %i WHERE option_name = %s LIMIT 1', $table, $lock_key ) );
Expand All @@ -290,7 +294,7 @@ private function acquire_lock( \wpdb $db, string $table, string $lock_key ): str
if ( 1 === $replaced ) {
return $lock_value;
}
if ( false === $replaced ) {
if ( false === $replaced && ! $this->is_transient_lock_contention( $db ) ) {
throw new WP_Agent_Run_Control_Store_Exception( 'Atomic run-control lock takeover failed.' );
}
}
Expand All @@ -301,6 +305,19 @@ private function acquire_lock( \wpdb $db, string $table, string $lock_key ): str
throw new WP_Agent_Run_Control_Store_Exception( 'Atomic run-control lock acquisition timed out.' );
}

/**
* Two connections racing an `INSERT IGNORE` (or a conditional takeover
* `UPDATE`) against the same unique lock row can make InnoDB report a
* deadlock or a lock-wait timeout instead of the clean "0 rows
* affected" outcome this loop already retries. Both are transient
* admission-race artifacts of the exact contention this method exists
* to serialize, not a genuine query or storage failure, so they get
* the same retry treatment as a lost duplicate-key race.
*/
private function is_transient_lock_contention( \wpdb $db ): bool {
return str_contains( $db->last_error, 'Deadlock found' ) || str_contains( $db->last_error, 'Lock wait timeout' );
}

private function release_lock( \wpdb $db, string $table, string $lock_key, string $lock_value ): void {
$released = $this->query( $db, $db->prepare( 'DELETE FROM %i WHERE option_name = %s AND option_value = %s', $table, $lock_key, $lock_value ) );
if ( false === $released ) {
Expand Down
Loading