DataTerminology.comNEW
A concurrency failure where transactions or processes wait indefinitely on each other’s locks, preventing progress.
DESCRIPTION
Deadlock occurs when two or more transactions hold locks that the others need, causing circular waiting. Databases detect deadlocks via wait-for graphs or timeouts and abort one transaction to break the cycle. Common causes include inconsistent lock ordering, large transaction scopes, and exclusive locks on contended resources. Prevention strategies include consistent lock ordering, shorter transactions, appropriate isolation levels, and careful indexing. Monitoring and logging deadlock events are essential to diagnose and remediate issues.
ALTERNATIVE DEFINITIONS
Wikipedia
In concurrent computing, a deadlock is a situation in which two or more competing actions are waiting for the other to finish, and thus neither ever does.
General computing framing beyond databases.
Microsoft Learn
A deadlock occurs when two or more sessions block each other because each session holds a lock on a resource that the other sessions are trying to lock.
Database-specific description with sessions and resources.
PostgreSQL Documentation
A deadlock is a situation where two or more transactions wait for each other, preventing any of them from proceeding.
Focuses on transactional waiting and resolution.
RELATED TERMS
Locking Two-Phase Locking Wait-For Graph Starvation Timeout Isolation Levels Transaction Isolation (ACID) Deadlock Detection
COMPETING TERMS
Livelock
KNOWLEDGE AREAS
Data Storage & Operations Data Engineering Software Engineering Infrastructure
TARGET AUDIENCE
Software Developer Data Engineer DevOps Engineer IT Manager Cloud Architect
EXTERNAL SOURCES
- Wikipedia - Deadlock
- Microsoft - Deadlock section in SQL Server
- PostgreSQL - Explicit locking and deadlocks
- MySQL - InnoDB deadlocks guide
TOP QUESTIONS
What is a Deadlock in databases?
It’s a circular wait where transactions lock resources needed by each other, preventing progress. Databases detect and resolve by aborting one transaction.
Category: Overview
What causes Deadlocks most often?
Inconsistent lock ordering, long transactions, and high contention on exclusive locks. Poor indexing can exacerbate locking conflicts.
Category: Mistakes
How can Deadlocks be prevented?
Use consistent lock ordering, keep transactions short, choose appropriate isolation levels, and reduce contention with indexing and batching.
Category: HowTo
What is the risk of Deadlocks in production?
They cause aborted transactions, user-facing errors, and performance degradation. Frequent deadlocks indicate design or operational issues.
Category: Risk
How are Deadlocks detected?
Wait-for graphs and periodic checks identify cycles. Timeouts provide a fallback. Logs help pinpoint offending statements.
Category: Management
What is the difference between Deadlock and Livelock?
In a livelock, processes change states without progress; in a deadlock, they remain blocked waiting for each other.
Category: Misunderstanding
What hot topics exist around Deadlocks?
Optimistic concurrency and lock-free structures reduce deadlock frequency, while MVCC improves read-write concurrency.
Category: Topic
How should applications handle Deadlocks?
Catch exceptions, implement retries with backoff, and instrument telemetry to analyze lock contention patterns.
Category: Requirements
Do higher isolation levels affect Deadlocks?
Yes, strict levels can increase contention and deadlock likelihood. Balance isolation with performance needs.
Category: Opinion