Aurora MySQL 在没有维护任务的情况下突然重启了

Question

Aurora MySQL 在没有维护任务的情况下突然重启了,错误日志显示 
```
2025-06-09T17:10:43Z UTC - mysqld got signal 6 ;
Most likely, you have hit a bug, but this error can also be caused by malfunctioning hardware.
BuildID[sha1]=82a011aba84a660d4bf03c9af2fb1fe3233b0bc1
Thread pointer: 0x40009b982000
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 400024748aa7 thread_stack 0x40000
/rdsdbbin/oscar/bin/mysqld(my_print_stacktrace(unsigned char const*, unsigned long)+0x30) [0x2647ff0]
/rdsdbbin/oscar/bin/mysqld(print_fatal_signal(int)+0x31c) [0x124d4dc]
/rdsdbbin/oscar/bin/mysqld(my_server_abort()+0x98) [0x124da98]
/rdsdbbin/oscar/bin/mysqld(my_abort()+0x14) [0x263ee14]
/rdsdbbin/oscar/bin/mysqld(ut_dbg_assertion_failed(char const*, char const*, unsigned long)+0x2a4) [0x2a0a224]
/rdsdbbin/oscar/bin/mysqld(ib::fatal_srcloc::~fatal_srcloc()+0xdc) [0x2a0f29c]
/rdsdbbin/oscar/bin/mysqld() [0x294c5ec]
/rdsdbbin/oscar/bin/mysqld(rec_offs_nth_extern(dict_index_t const*, unsigned long const*, unsigned long)+0x6c) [0x294c88c]
/rdsdbbin/oscar/bin/mysqld(row_sel_store_mysql_field(unsigned char*, row_prebuilt_t*, unsigned char const*, dict_index_t const*, dict_index_t const*, unsigned long const*, unsigned long, mysql_row_templ_t const*, unsigned long, lob::undo_vers_t*, mem_block_info_t*&)+0x58) [0x298dcb8]
/rdsdbbin/oscar/bin/mysqld(row_sel_store_mysql_rec(unsigned char*, row_prebuilt_t*, unsigned char const*, dtuple_t const*, bool, dict_index_t const*, dict_index_t const*, unsigned long const*, bool, lob::undo_vers_t*, mem_block_info_t*&)+0x340) [0x2990b40]
/rdsdbbin/oscar/bin/mysqld(row_search_mvcc(unsigned char*, page_cur_mode_t, row_prebuilt_t*, unsigned long, unsigned long)+0x1c84) [0x29958c4]
/rdsdbbin/oscar/bin/mysqld(ha_innobase::index_read(unsigned char*, unsigned char const*, unsigned int, ha_rkey_function)+0x378) [0x2887c98]
/rdsdbbin/oscar/bin/mysqld(handler::ha_index_read_map(unsigned char*, unsigned char const*, unsigned long, ha_rkey_function)+0x2bc) [0x1421e7c]
/rdsdbbin/oscar/bin/mysqld(handler::read_range_first(key_range const*, key_range const*, bool, bool)+0x60) [0x1422ac0]
/rdsdbbin/oscar/bin/mysqld(ha_innobase::read_range_first(key_range const*, key_range const*, bool, bool)+0x1c) [0x28553dc]
/rdsdbbin/oscar/bin/mysqld(handler::multi_range_read_next(char**)+0x178) [0x1423598]
/rdsdbbin/oscar/bin/mysqld(handler::ha_multi_range_read_next(char**)+0x184) [0x1423d44]
/rdsdbbin/oscar/bin/mysqld(IndexRangeScanIterator::Read()+0x48) [0x1678fc8]
/rdsdbbin/oscar/bin/mysqld(Sql_cmd_delete::delete_from_single_table(THD*)+0x950) [0x16e7e30]
/rdsdbbin/oscar/bin/mysqld(Sql_cmd_dml::execute(THD*)+0x158) [0x10585b8]
/rdsdbbin/oscar/bin/mysqld(mysql_execute_command_internal(THD*, bool)+0x122c) [0xffc3cc]
/rdsdbbin/oscar/bin/mysqld(mysql_execute_command(THD*, bool)+0x10c) [0xfff5cc]
/rdsdbbin/oscar/bin/mysqld(dispatch_sql_command(THD*, Parser_state*)+0x34c) [0x100036c]
/rdsdbbin/oscar/bin/mysqld(dispatch_command(THD*, COM_DATA const*, enum_server_command)+0xcbc) [0x100165c]
/rdsdbbin/oscar/bin/mysqld(do_command(THD*)+0x1e0) [0x10030e0]
/rdsdbbin/oscar/bin/mysqld(THD_task::process_connection()+0x1c4) [0x123bfc4]
/rdsdbbin/oscar/bin/mysqld(Thread_pool::worker_loop()+0x174) [0x1239a14]
/rdsdbbin/oscar/bin/mysqld(Thread_pool::worker_launch(void*)+0x24) [0x1239c24]
/rdsdbbin/oscar/bin/mysqld() [0x2b89fd0]
/lib64/libpthread.so.0(+0x7230) [0x40001bb96230]
/lib64/libc.so.6(+0xdb7dc) [0x40001bf8e7dc]
```

当重启时，我的rds  CPU使用率大概在60%，内存在80%左右

Zhao Peng · Answer

以下是 Aurora MySQL 在无维护任务下突然重启的可能性原因，请参考

---

### ✅ 主要原因总结：

从错误日志来看，这是由于 **Aurora MySQL 内部 bug 引起的崩溃（signal 6）**。具体表现为：

* MySQL 在执行 `DELETE` 操作时，InnoDB 存储引擎内部出现了 **断言失败（assertion failed）**；
* 断言失败表示系统发现了“绝对不应该出现的异常情况”，于是主动触发了 **异常终止（abort）**；
* 日志中还涉及到 `LOB`（大型对象）处理逻辑，可能和某个字段使用了 `TEXT` 或 `BLOB` 类型有关；
* 这并非人为操作或配置问题，也不是你方服务器负载的问题，而是**Aurora 自身引起的系统级错误**。

---

### 🔍 简单解释：

* **signal 6（SIGABRT）** 是 Linux 系统中进程自己触发的致命错误信号，常见于程序遇到重大 bug 或非法状态；
* Aurora MySQL 为了防止数据进一步损坏，在发现问题后自动进行了重启。

---

### 🛠 推荐你采取的措施：

1. **联系 AWS 官方支持**
   附上错误日志与重启时间，让他们确认是否为已知 bug，或是否需要打补丁或升级版本。

2. **确认 Aurora MySQL 的版本**
   如果版本较老，建议升级。Aurora 会定期发布 bug 修复版。

3. **检查触发崩溃的 SQL 语句和表结构**
   查看是否是 `DELETE` 操作触发崩溃，确认是否涉及 `BLOB/TEXT` 类型，是否存在历史数据损坏可能。

4. **启用性能监控（Performance Insights）与增强监控（Enhanced Monitoring）**
   可帮助日后排查类似问题。

---

### ⚠️ 注意事项：

* 本次 CPU 60%、内存 80% 使用率属于正常范围，不是重启原因；
* 此类 bug 不常见，但一旦发生就可能导致数据库自动重启。

re:Post Agent · Answer

The error log you've shared indicates that your Aurora MySQL instance crashed due to a signal 6 (SIGABRT), which typically suggests the database encountered a critical error that forced it to abort. This appears to be related to an InnoDB engine issue during a DELETE operation, as shown in the stack trace.

Looking at the backtrace, the crash occurred during a database operation that involved index reading and row selection. Specifically, the error happened in the InnoDB storage engine while processing a DELETE statement, as indicated by the "Sql_cmd_delete::delete_from_single_table" entry in the stack trace.

This type of crash is often caused by one of the following:

1. A bug in the MySQL database engine
2. Data corruption issues
3. Memory-related problems

When Aurora MySQL encounters such a critical error, it will automatically restart to recover from the crash. This is part of Aurora's self-healing capabilities.

To address this issue, you could:

1. Check for any recent heavy DELETE operations that might have triggered this crash
2. Consider running a consistency check on your tables to identify potential data corruption
3. Monitor for recurrence of the issue and collect more detailed logs
4. If the issue persists, you might need to consider upgrading your Aurora MySQL version, as this could be a known bug that's fixed in a newer release

If this happens repeatedly, you should contact AWS Support with the error logs for further investigation, as they can provide more specific guidance based on the exact nature of the crash.  
**Sources**  
[postgreSQL aurora serverless v1 upgrade from 11.21 to 13.12 | AWS re:Post](https://repost.aws/questions/QUSV6dB340SR-FoF2_pvu4-A/postgresql-aurora-serverless-v1-upgrade-from-11-21-to-13-12)  
[AWS RDS PostgreSQL Upgrade 15.4 -> 16.2 Fails | AWS re:Post](https://repost.aws/questions/QUxaWRyKXRRSqlN7w_O12kvw/aws-rds-postgresql-upgrade-15-4-16-2-fails)

Aurora MySQL 在没有维护任务的情况下突然重启了

✅ 主要原因总结：

🔍 简单解释：

🛠 推荐你采取的措施：

⚠️ 注意事项：

添加回答

相关内容