<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:media="http://search.yahoo.com/mrss/" version="2.0"><channel><title>梦不止于想(	学习编码集)</title><link>https://blog.codealy.top</link><atom:link href="https://blog.codealy.top/rss.xml" rel="self" type="application/rss+xml"/><description>Easy to understand and humorous</description><generator>Halo v2.22.2</generator><language>zh-cn</language><image><url>https://s.codealy.com/favicon.ico</url><title>梦不止于想(	学习编码集)</title><link>https://blog.codealy.top</link></image><lastBuildDate>Thu, 9 Apr 2026 08:14:14 GMT</lastBuildDate><item><title><![CDATA[WEB服务的API接口的安全措施概述]]></title><link>https://blog.codealy.top/archives/web_api_safety</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=WEB%E6%9C%8D%E5%8A%A1%E7%9A%84API%E6%8E%A5%E5%8F%A3%E7%9A%84%E5%AE%89%E5%85%A8%E6%8E%AA%E6%96%BD%E6%A6%82%E8%BF%B0&amp;url=/archives/web_api_safety" width="1" height="1" alt="" style="opacity:0;">本文系统梳理了Web服务API安全的核心措施，涵盖传输层（HTTPS加密、HSTS）、身份认证（OAuth 2.0、WebAuthn）、请求防护（签名验证、输入过滤、防重放）、访问控制（IP黑白名单、多维度限流）、可观测性（结构化日志、幂等设计）、数据隐私（脱敏策略）及架构级防护（API网关集中治理）。强调纵深防御体系需贯穿开发全周期，结合最小权限原则、持续监控与合规要求，构建动态迭代的安全框架，以抵御数据泄露、未授权访问等风险。]]></description><guid isPermaLink="false">/archives/web_api_safety</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fjava%2Fsecurity%2Fapi_safety.png&amp;size=m" type="image/jpeg" length="1039345"/><category>Java</category><pubDate>Fri, 26 Dec 2025 05:37:37 GMT</pubDate></item><item><title><![CDATA[MYSQL索引失效常见场景 - 数据库性能优化]]></title><link>https://blog.codealy.top/archives/mysql_index_failure_scenarios</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=MYSQL%E7%B4%A2%E5%BC%95%E5%A4%B1%E6%95%88%E5%B8%B8%E8%A7%81%E5%9C%BA%E6%99%AF%20-%20%E6%95%B0%E6%8D%AE%E5%BA%93%E6%80%A7%E8%83%BD%E4%BC%98%E5%8C%96&amp;url=/archives/mysql_index_failure_scenarios" width="1" height="1" alt="" style="opacity:0;">本文深入剖析MySQL索引失效的六大常见场景：表达式计算、数据类型不匹配、模糊查询前缀问题、复合索引最左匹配原则、OR条件查询及负向查询。从B+树索引原理出发，解释失效机制并提供针对性解决方案，如改写计算条件、确保类型一致、优化LIKE模式、设计复合索引顺序、使用UNION替代OR等。强调通过执行计划分析验证索引效果，遵循“纯净索引列”“类型严格匹配”“覆盖索引”等原则，结合持续监控与工具辅助，构建高效数据访问层，系统性提升数据库查询性能。]]></description><guid isPermaLink="false">/archives/mysql_index_failure_scenarios</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatabase%2Fmysql%2Findex%2Fmysql_index_failure_scenarios.png&amp;size=m" type="image/jpeg" length="1346139"/><category>MYSQL</category><pubDate>Wed, 24 Dec 2025 04:52:00 GMT</pubDate></item><item><title><![CDATA[Spring循环依赖解析与三级缓存机制]]></title><link>https://blog.codealy.top/archives/spring_loop_depend_on</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Spring%E5%BE%AA%E7%8E%AF%E4%BE%9D%E8%B5%96%E8%A7%A3%E6%9E%90%E4%B8%8E%E4%B8%89%E7%BA%A7%E7%BC%93%E5%AD%98%E6%9C%BA%E5%88%B6&amp;url=/archives/spring_loop_depend_on" width="1" height="1" alt="" style="opacity:0;">本文深入剖析Spring框架解决循环依赖的三级缓存机制。该机制通过singletonObjects（完全初始化Bean）、earlySingletonObjects（半成品Bean）和singletonFactories（对象工厂）三级缓存协同工作：实例化时将半成品工厂存入三级缓存，依赖注入时若发现循环依赖则提前暴露半成品引用至二级缓存，完成初始化后升级至一级缓存。此机制支持字段/Setter注入的循环依赖，但无法解决构造器注入及原型Bean的循环依赖。文章强调开发者应理解机制而非依赖它，建议通过事件驱动、门面模式等设计优化架构，避免循环依赖的产生。]]></description><guid isPermaLink="false">/archives/spring_loop_depend_on</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fjava%2Fspring%2FDI%2Fspring_loop_depend.png&amp;size=m" type="image/jpeg" length="554268"/><category>Java</category><pubDate>Tue, 23 Dec 2025 02:34:26 GMT</pubDate></item><item><title><![CDATA[SQL查询语句执行顺序深度解析：从语法到执行的完整旅程]]></title><link>https://blog.codealy.top/archives/sql_execution_order</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=SQL%E6%9F%A5%E8%AF%A2%E8%AF%AD%E5%8F%A5%E6%89%A7%E8%A1%8C%E9%A1%BA%E5%BA%8F%E6%B7%B1%E5%BA%A6%E8%A7%A3%E6%9E%90%EF%BC%9A%E4%BB%8E%E8%AF%AD%E6%B3%95%E5%88%B0%E6%89%A7%E8%A1%8C%E7%9A%84%E5%AE%8C%E6%95%B4%E6%97%85%E7%A8%8B&amp;url=/archives/sql_execution_order" width="1" height="1" alt="" style="opacity:0;">理解SQL执行顺序对高效数据库查询至关重要。SQL实际执行顺序并非按书写顺序，而是从FROM/JOIN开始，依次经过ON、WHERE、GROUP BY、HAVING、SELECT、DISTINCT、ORDER BY，最后到LIMIT/OFFSET。掌握此顺序可优化性能，如尽早过滤数据减少处理量、合理使用索引加速查询，并通过执行计划分析瓶颈，从而编写更高效的SQL语句。]]></description><guid isPermaLink="false">/archives/sql_execution_order</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatabase%2Fmysql%2Fsql-execution-order.jpg&amp;size=m" type="image/jpeg" length="373494"/><category>数据库</category><pubDate>Thu, 11 Dec 2025 02:32:59 GMT</pubDate></item><item><title><![CDATA[Spring 事务失效的八大场景深度解析]]></title><link>https://blog.codealy.top/archives/spring_tx_lose</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Spring%20%E4%BA%8B%E5%8A%A1%E5%A4%B1%E6%95%88%E7%9A%84%E5%85%AB%E5%A4%A7%E5%9C%BA%E6%99%AF%E6%B7%B1%E5%BA%A6%E8%A7%A3%E6%9E%90&amp;url=/archives/spring_tx_lose" width="1" height="1" alt="" style="opacity:0;">本文系统分析了Spring事务失效的常见原因及解决方案。核心问题包括：代理机制失效（非public方法、内部调用、final/static方法）、异常处理不当（异常类型不匹配、异常被吞没）、配置错误（未启用事务管理、超时设置过短、传播机制错误）、数据库层面问题（引擎不支持事务）、并发场景冲突（@Async与@Transactional冲突）。解决需关注方法可见性、异常抛出、事务传播行为配置，并通过调试日志和编程式事务排查问题。理解Spring AOP机制和事务原理是避免失效的关键。]]></description><guid isPermaLink="false">/archives/spring_tx_lose</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fjava%2Fspring%2Ftx%2Fspring-tx-lose.jpg&amp;size=m" type="image/jpeg" length="404292"/><category>Java</category><pubDate>Wed, 10 Dec 2025 08:31:08 GMT</pubDate></item><item><title><![CDATA[数据库安全网关体系中，对Long、Date类型列加密的方法]]></title><link>https://blog.codealy.top/archives/fpe</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=%E6%95%B0%E6%8D%AE%E5%BA%93%E5%AE%89%E5%85%A8%E7%BD%91%E5%85%B3%E4%BD%93%E7%B3%BB%E4%B8%AD%EF%BC%8C%E5%AF%B9Long%E3%80%81Date%E7%B1%BB%E5%9E%8B%E5%88%97%E5%8A%A0%E5%AF%86%E7%9A%84%E6%96%B9%E6%B3%95&amp;url=/archives/fpe" width="1" height="1" alt="" style="opacity:0;">本文探讨数据库安全网关中Long与Date类型数据的加密难题，核心在于平衡安全性与业务查询需求。传统加密（如AES/SM4）会破坏数据类型特性，导致查询功能丧失。格式保留加密（FPE）作为关键解决方案，通过有限域映射原理，使加密后数据保持原始格式与长度（如Long仍为数值，Date仍为日期格式），从而支持等值查询和部分范围查询，完美兼容现有数据库模式。但FPE实现复杂、性能较传统加密低，且安全性审查尚需加强。]]></description><guid isPermaLink="false">/archives/fpe</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Ffile.codealy.top%2Fd%2Fdisk%2Fhalo%2Fthumbnail%2FCatalanPyrenees_1920x1080.jpg%3Fsign%3DwaSA7Fvs6iT433aTvBhnvw4VFsAH6idV-pu3PIJ0sEI%3D%3A0&amp;size=m" type="image/jpeg" length="337820"/><category>数据库</category><pubDate>Tue, 28 Oct 2025 11:19:36 GMT</pubDate></item><item><title><![CDATA[浅谈常见的八类数据库加密技术]]></title><link>https://blog.codealy.top/archives/database_enc_style</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=%E6%B5%85%E8%B0%88%E5%B8%B8%E8%A7%81%E7%9A%84%E5%85%AB%E7%B1%BB%E6%95%B0%E6%8D%AE%E5%BA%93%E5%8A%A0%E5%AF%86%E6%8A%80%E6%9C%AF&amp;url=/archives/database_enc_style" width="1" height="1" alt="" style="opacity:0;">本文概述了加密技术从通信安全向数据全生命周期保护的范式转变。面对大数据时代数据动态流转的挑战，传统边界防御已不足，需转向“保护数据本身”的安全理念。文章深入剖析八种主流加密技术（应用内加密、数据库网关、触发器+视图、TDE、UDF、加密驱动、TFE、FDE），涵盖应用层、数据库层及文件系统层的部署方案，对比其加密粒度、性能、防DBA能力及实施成本。核心结论是：单一技术无法满足所有场景，需根据数据状态（静态/动态/使用中）与业务需求，组合构建纵深防御体系，实现数据流转中的持续安全保护。]]></description><guid isPermaLink="false">/archives/database_enc_style</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fenc%2Fdatabase_enc_in_gateway_sql.png&amp;size=m" type="image/jpeg" length="41806"/><category>数据库</category><pubDate>Wed, 22 Oct 2025 03:29:49 GMT</pubDate></item><item><title><![CDATA[Doris SQL解析]]></title><link>https://blog.codealy.top/archives/doris_sql_parser</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Doris%20SQL%E8%A7%A3%E6%9E%90&amp;url=/archives/doris_sql_parser" width="1" height="1" alt="" style="opacity:0;">Doris SQL解析包含五个核心步骤：1）词法语法分析生成AST（采用jflex/java cup技术）；2）语义分析与重写（元信息解析、合法性检查、常量折叠、谓词转join等RBO优化）；3）生成单机逻辑计划（构建算子树，进行投影/谓词下推、分区裁剪、Join重排序等优化）；4）生成分布式计划（拆分PlanFragment树，支持broadcast/hash partition/colocate/bucket shuffle四种join算法）；5）物理执行计划调度（分配BE节点、选择副本、实现并发执行）。整个过程通过递归优化和规则重写，最大化并行度与数据本地化。]]></description><guid isPermaLink="false">/archives/doris_sql_parser</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fdoris%2FDoris_SQL_parser.webp&amp;size=m" type="image/jpeg" length="55456"/><category>Doris</category><pubDate>Sun, 19 Oct 2025 05:01:00 GMT</pubDate></item><item><title><![CDATA[Doris]]></title><link>https://blog.codealy.top/archives/doris</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Doris&amp;url=/archives/doris" width="1" height="1" alt="" style="opacity:0;">Doris架构分FE（前端）、BE（后端）和存储层：FE负责查询解析、元数据管理，含Leader/Follower保高可用、Observer扩展查询；BE存储数据并分布式执行查询，通过副本保可靠；存储层用列式存储，数据分Tablet存于BE。数据模型有三类：Aggregate按key聚合适合报表汇总；Uniq主键模型保证唯一，分读时/写时合并；Duplicate允许key重复保留明细，适用Ad-hoc查询。选型需根据聚合需求、主键约束等场景确定。]]></description><guid isPermaLink="false">/archives/doris</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fdoris%2FDoris_single_plan.webp&amp;size=m" type="image/jpeg" length="23008"/><category>Doris</category><pubDate>Thu, 16 Oct 2025 04:58:00 GMT</pubDate></item><item><title><![CDATA[Calcite 元数据定义]]></title><link>https://blog.codealy.top/archives/calcite_meta</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Calcite%20%E5%85%83%E6%95%B0%E6%8D%AE%E5%AE%9A%E4%B9%89&amp;url=/archives/calcite_meta" width="1" height="1" alt="" style="opacity:0;">Calcite执行流程的核心包括三个部分：元数据定义、优化规则管理和最优计划执行。元数据用于校验SqlNode语法树并为CBO优化提供统计信息，如示例中的JSON文件定义；优化规则被优化器使用以改写逻辑计划并生成最优执行计划；执行器基于最优计划在不同存储引擎上进行执行，确保高效数据处理。]]></description><guid isPermaLink="false">/archives/calcite_meta</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fcalcite%2Fcalcite_core.png&amp;size=m" type="image/jpeg" length="77420"/><category>Calcite</category><pubDate>Mon, 13 Oct 2025 04:55:00 GMT</pubDate></item><item><title><![CDATA[Apache Calcite：构建统一数据查询的基石]]></title><link>https://blog.codealy.top/archives/calcite</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Apache%20Calcite%EF%BC%9A%E6%9E%84%E5%BB%BA%E7%BB%9F%E4%B8%80%E6%95%B0%E6%8D%AE%E6%9F%A5%E8%AF%A2%E7%9A%84%E5%9F%BA%E7%9F%B3&amp;url=/archives/calcite" width="1" height="1" alt="" style="opacity:0;">Apache Calcite 是开源动态数据管理框架，旨在解决大数据时代的数据孤岛与异构数据源整合挑战。其核心架构以查询优化器为中心，通过 SQL 解析器、验证器将语句转为逻辑关系代数表达式（RelNode），结合 RBO（200+ 优化规则）与 CBO（基于统计信息）生成最优执行计划。框架采用“逻辑与物理分离”设计，支持 JDBC/Java API 调用，并通过可插拔适配器连接 Hive、Elasticsearch 等数据源，提供统一查询层。其模块化架构（如元数据管理、表达式构建）及 CSV 示例验证了跨源查询的灵活性与扩展性。]]></description><guid isPermaLink="false">/archives/calcite</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fcalcite%2Fcalcite.png&amp;size=m" type="image/jpeg" length="143414"/><category>Calcite</category><pubDate>Sat, 11 Oct 2025 08:53:00 GMT</pubDate></item><item><title><![CDATA[RSA加解密流程]]></title><link>https://blog.codealy.top/archives/asym_rsa</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=RSA%E5%8A%A0%E8%A7%A3%E5%AF%86%E6%B5%81%E7%A8%8B&amp;url=/archives/asym_rsa" width="1" height="1" alt="" style="opacity:0;">RSA是一种非对称加密算法，通过公钥加密、私钥解密实现安全通信，无需直接传递密钥。其安全性基于大整数因数分解的计算困难性。算法核心依赖数论基础：素数定义、模运算与同余关系、互质条件及欧拉函数φ(n)（计算与n互质的整数数量）。欧拉定理（a^φ(n) ≡ 1 mod n）和模反元素（ab ≡ 1 mod n）构成RSA的数学原理，确保加解密的可行性。私钥需保密，公钥可公开。此外，文章提及RSA已知高位攻击利用Coppersmith定理，在部分密钥高位泄露时可破解加密。]]></description><guid isPermaLink="false">/archives/asym_rsa</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2FWebSecurity%2Finterview%2FRSA%2FRSA-know-why.png&amp;size=m" type="image/jpeg" length="48155"/><category>算法</category><pubDate>Thu, 9 Oct 2025 03:13:00 GMT</pubDate></item><item><title><![CDATA[XSS漏洞]]></title><link>https://blog.codealy.top/archives/xss</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=XSS%E6%BC%8F%E6%B4%9E&amp;url=/archives/xss" width="1" height="1" alt="" style="opacity:0;">XSS（跨站脚本攻击）是通过在网页注入恶意脚本，当用户访问时执行窃取数据或实施攻击的漏洞。静态站点不受影响。攻击主要分三类：反射型（非持久，通过URL参数反射执行）、存储型（持久，恶意代码存于服务器）、DOM型（客户端直接操作DOM执行）。危害包括盗取账号、控制数据、篡改页面、挂马等。防御需综合措施：输入过滤、输出转义、设置HttpOnly（防JS读取Cookie）、指定字符集、验证跳转URL等。]]></description><guid isPermaLink="false">/archives/xss</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2FWebSecurity%2Floophole%2Fxss%2Fhtml_dom.png&amp;size=m" type="image/jpeg" length="40177"/><category>网络安全</category><pubDate>Tue, 7 Oct 2025 08:06:00 GMT</pubDate></item><item><title><![CDATA[CSRF漏洞]]></title><link>https://blog.codealy.top/archives/csrf</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=CSRF%E6%BC%8F%E6%B4%9E&amp;url=/archives/csrf" width="1" height="1" alt="" style="opacity:0;">CSRF（跨站请求伪造）是攻击者利用用户已登录信任网站的Cookie，伪装其身份向目标网站发送恶意请求的攻击。其成功关键在于请求参数可被预测，且用户在未退出信任网站时访问了危险网站。攻击分GET型（通过`&lt;img&gt;`等标签发起）和POST型（利用表单或JS自动提交）。防御措施包括：添加验证码强制用户交互；在请求中嵌入随机Token并验证（URL参数或HTTP头）；校验HTTP Referer字段确保请求来源合法。其中Token机制是核心防御手段，需保证其随机性与保密性。]]></description><guid isPermaLink="false">/archives/csrf</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2FWebSecurity%2Floophole%2Fcsrf_refer_view.gif&amp;size=m" type="image/jpeg" length="1144578"/><category>网络安全</category><pubDate>Sun, 28 Sep 2025 08:04:00 GMT</pubDate></item><item><title><![CDATA[TLS 如何防止中间人攻击？]]></title><link>https://blog.codealy.top/archives/tls</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=TLS%20%E5%A6%82%E4%BD%95%E9%98%B2%E6%AD%A2%E4%B8%AD%E9%97%B4%E4%BA%BA%E6%94%BB%E5%87%BB%EF%BC%9F&amp;url=/archives/tls" width="1" height="1" alt="" style="opacity:0;">TLS通过加密和身份验证机制有效防御中间人攻击。其结合对称与非对称加密技术，在握手过程中验证服务器数字证书，确保通信双方身份真实性，并协商生成临时会话密钥。完美前向保密（PFS）功能使每个会话使用独立密钥，即使私钥泄露也无法解密历史通信。Diffie-Hellman等安全密钥交换方法进一步保障密钥协商过程不被窃听，从而实现数据机密性、完整性和身份认证，阻断攻击者拦截或篡改通信的路径。]]></description><guid isPermaLink="false">/archives/tls</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fcn.bing.com%2Fth%3Fid%3DOHR.ArambolBeach_ZH-CN2149857876_1920x1080.jpg&amp;size=m" type="image/jpeg" length="344547"/><category>网络安全</category><pubDate>Thu, 25 Sep 2025 08:03:00 GMT</pubDate></item><item><title><![CDATA[Clickhouse]]></title><link>https://blog.codealy.top/archives/clickhouse</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Clickhouse&amp;url=/archives/clickhouse" width="1" height="1" alt="" style="opacity:0;">ClickHouse是一款面向OLAP场景的列式数据库，支持完整SQL操作、列式存储与压缩、向量化执行引擎及并行处理，通过分片（Shard）和副本（Node）实现分布式架构。其依赖ZooKeeper进行元数据管理和分布式协调，但存在性能瓶颈与运维复杂性问题。主要限制包括不支持事务、二级索引及窗口函数，行删除效率低，且聚合结果需单机内存容纳。适用于读多写少、大批次更新、宽表查询、高吞吐量分析等场景，尤其适合单次查询涉及大表过滤聚合、结果量小的分析任务。]]></description><guid isPermaLink="false">/archives/clickhouse</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fdatahouse%2Fclickhouse%2FClickHouse_framework.jpg&amp;size=m" type="image/jpeg" length="48385"/><category>Clickhouse</category><pubDate>Sun, 21 Sep 2025 09:35:00 GMT</pubDate></item><item><title><![CDATA[MYSQL - InnoDB Difficult to find free blocks in the buffer pool]]></title><link>https://blog.codealy.top/archives/mysql_innodb</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=MYSQL%20-%20InnoDB%20Difficult%20to%20find%20free%20blocks%20in%20the%20buffer%20pool&amp;url=/archives/mysql_innodb" width="1" height="1" alt="" style="opacity:0;">在16G 8C服务器上运行MySQL 5.7时，大数据量聚合查询导致系统崩溃，无法连接或停止。日志显示InnoDB缓冲池难以找到空闲块，刷新失败次数超百万，建议增大缓冲池大小或升级OS。解决方法：通过top定位进程并kill，修改my.cnf设置innodb_buffer_pool_size=12G（总内存80%）和innodb_buffer_pool_instances=6，重启MySQL。针对page_cleaner循环超时警告，执行SET GLOBAL innodb_lru_scan_depth=256优化。]]></description><guid isPermaLink="false">/archives/mysql_innodb</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fcn.bing.com%2Fth%3Fid%3DOHR.SquirrelNesting_ZH-CN7673817247_1920x1080.jpg&amp;size=m" type="image/jpeg" length="344536"/><category>MYSQL</category><pubDate>Tue, 16 Sep 2025 08:26:00 GMT</pubDate></item><item><title><![CDATA[什么是zxid]]></title><link>https://blog.codealy.top/archives/zxid</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=%E4%BB%80%E4%B9%88%E6%98%AFzxid&amp;url=/archives/zxid" width="1" height="1" alt="" style="opacity:0;">zxid是ZooKeeper中事务的全局唯一ID，由高32位epoch（选举周期）和低32位counter（事务计数）组成。当counter超过42亿时，epoch加一，可能导致事务ID冲突，破坏全序关系，因此ZooKeeper主动触发选主。选主对客户端通常无感知，但依赖Disconnected事件的应用（如LeaderLatch）可能受影响。为避免溢出，需监控zxid并在接近溢出时手动选主。选主和重启一般不导致session过期。]]></description><guid isPermaLink="false">/archives/zxid</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2Fzookeeper%2Fzk_zxid.png&amp;size=m" type="image/jpeg" length="71911"/><category>Zookeeper</category><pubDate>Mon, 8 Sep 2025 08:21:00 GMT</pubDate></item><item><title><![CDATA[Rust]]></title><link>https://blog.codealy.top/archives/rust</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Rust&amp;url=/archives/rust" width="1" height="1" alt="" style="opacity:0;">以下是文章摘要（148字）： Rust通过rustup工具管理安装与版本迭代，支持多平台及Beta/Nightly等版本，升级和卸载命令分别为`rustup update`和`rustup self uninstall`。宏作为元编程工具，分声明式宏（基于文本替换）和过程宏（操作语法树，含派生/属性/函数式三类）。常见宏如`println!`、`vec!`等，与函数的核心区别在于宏在编译时展开，可接受任意参数并生成代码，灵活性更高。]]></description><guid isPermaLink="false">/archives/rust</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fcn.bing.com%2Fth%3Fid%3DOHR.SarekSweden_ZH-CN9728518595_1920x1080.jpg&amp;size=m" type="image/jpeg" length="325741"/><category>Rust</category><pubDate>Fri, 5 Sep 2025 05:46:00 GMT</pubDate></item><item><title><![CDATA[Redis 概述]]></title><link>https://blog.codealy.top/archives/redis</link><description><![CDATA[<img src="https://blog.codealy.top/plugins/feed/assets/telemetry.gif?title=Redis%20%E6%A6%82%E8%BF%B0&amp;url=/archives/redis" width="1" height="1" alt="" style="opacity:0;">Redis是一个高性能、开源的key-value数据库，支持数据持久化、多种数据结构（如list、set、zset、hash）及主从备份。其快速性源于基于RAM的存储（比磁盘快1000倍）、IO复用与单线程执行循环，以及高效底层数据结构。Redis Cluster采用16384个槽位进行数据分片，选择此数而非65536，是为减少心跳包大小（2k vs 8k空间）并确保可扩展性，最多支持1000个主节点。]]></description><guid isPermaLink="false">/archives/redis</guid><dc:creator>fengyang</dc:creator><enclosure url="https://blog.codealy.top/apis/api.storage.halo.run/v1alpha1/thumbnails/-/via-uri?uri=https%3A%2F%2Fs.codealy.com%2Fwhosly.io%2FRedis%2FWhy_is_Redis_so_fast.png&amp;size=m" type="image/jpeg" length="367443"/><category>Redis</category><pubDate>Sun, 31 Aug 2025 10:47:00 GMT</pubDate></item></channel></rss>