0 views collected around this technical thread.
This article provides a comprehensive overview of database transactions, explaining ACID properties, transaction states, explicit and implicit transaction control, isolation levels, and how to configure and use them in MySQL with practical SQL examples and code snippets.
The article examines a MySQL InnoDB deadlock that occurs when concurrent inventory‑write transactions in a pre‑release environment contend for row and gap locks, explains InnoDB lock types, isolation‑level effects, and proposes solutions such as distributed locks and isolation level adjustments.
This article examines how common business operations using SELECT ... FOR UPDATE and INSERT can cause Next‑Key lock deadlocks in MySQL under repeatable‑read isolation, demonstrates the locking behavior with practical SQL examples, and proposes strategies such as post‑locking and primary‑key updates to prevent such deadlocks.
This article explains database isolation levels, including read committed, repeatable read, serializable, and snapshot isolation, discusses dirty reads, phantom reads, and how various systems such as SQL Server, PostgreSQL, MySQL, Oracle, MongoDB, and others implement or differ in these isolation semantics.
This article explains how MySQL handles concurrent transactions using MVCC, lock mechanisms, and various isolation levels, detailing dirty writes, dirty reads, non‑repeatable reads, phantom reads, and the internal structures such as ReadView that enable reliable data consistency.
This article explains the concept of phantom reads, distinguishes them from non‑repeatable reads, critiques the ANSI SQL isolation‑level definitions, and shows how MySQL implements repeatable‑read using locks and MVCC to prevent both strict and loosely defined phantom anomalies.
This article explains why MySQL chooses Repeatable Read as its default transaction isolation level, contrasting it with Oracle’s default Read Committed, and discusses how binlog formats, GAP locks, and historical bugs influence this design decision.
The article explains why an INSERT INTO SELECT statement can lock an entire InnoDB table in MySQL, demonstrates the lock behavior with transaction examples, and offers two solutions—adding an index on the filtered column and adjusting the transaction isolation level—to avoid table‑wide blocking.
This article explains the concept of database transactions, the ACID properties, MySQL's four isolation levels, the phenomena they cause such as dirty reads, non‑repeatable reads and phantom reads, and provides step‑by‑step experiments demonstrating each isolation level in practice.
This article explains the concepts of isolation levels, dirty reads, phantom rows, and how various relational and NoSQL databases implement these phenomena, providing practical guidance on choosing the appropriate isolation level to ensure data consistency and performance.
This article explains how to configure Spring transaction isolation levels and propagation behaviors, illustrates the effects of dirty read, non‑repeatable read, and phantom read with examples, and provides practical code solutions for handling rollback of failed sub‑tasks and self‑invocation issues.
This article explains the four transaction isolation levels, defines phantom reads, and details how MySQL's InnoDB engine uses MVCC snapshot reads and next‑key locking (including the next‑key principle and lock composition) to prevent phantom reads under the REPEATABLE READ isolation level.
This article explains the concepts of isolation levels, dirty reads, and phantom reads across various relational and NoSQL databases, describing how each system implements or lacks transaction isolation and the practical impact on data consistency and performance.
A detailed case study investigates why DBLE middleware achieves only about 70% of MyCat's performance, revealing that excessive sorting CPU usage, a cmp‑function inefficiency, and a delete‑loop bug under REPEATABLE_READ isolation cause the slowdown, and demonstrates how adjusting isolation levels and fixing the cmp implementation restores expected performance.
This article examines the challenges of using transactions with database read/write splitting, including potential master‑slave switching errors, connection‑pool deadlocks caused by isolation levels and propagation, and presents Spring‑based dynamic data source and transaction management solutions.
This article explains what the NOLOCK hint does in SQL Server, why it is used, how it relates to transaction isolation levels, presents practical tests, discusses the relevance of NOLOCK after newer SQL Server versions, and shows why MySQL does not need an equivalent hint, offering guidance on when to apply it.
This article explains Spring's declarative transaction mechanism, detailing transaction attributes such as propagation behavior, isolation levels, timeout and read‑only flags, and illustrates each propagation type with Java code examples to show how transactions are started, joined, suspended, or nested.