OpenShift OKD下MySQL持久化部署启动失败求助

看了你的日志，核心问题很明确——InnoDB的核心数据文件ibdata1损坏了，头部全是0字节，导致引擎初始化直接崩溃。临时存储能正常运行是因为每次都是全新的文件系统，没有残留问题，权限也都是容器默认配置好的，但持久化存储就容易踩权限或数据残留的坑。

下面给你几个靠谱的解决步骤，按顺序尝试：

第一步：检查并修复存储卷的权限 #

MySQL容器默认使用mysql用户运行（UID通常是27），如果你的持久化存储卷不是这个用户能读写的，初始化时写入文件就会出问题，最终生成损坏的文件。

操作步骤：

1. 先删除当前出问题的PersistentVolumeClaim（PVC）和对应的PersistentVolume（PV）（如果是静态PV的话）
2. 重新创建PVC后，启动一个调试容器挂载它修改权限：

   oc run debug-pod --image=busybox --command -- sleep 3600
   oc rsh debug-pod
   chown -R 27:27 /var/lib/mysql  # 注意：这里的路径要和MySQL容器里的挂载路径完全一致
   exit
   oc delete pod debug-pod
   

3. 重新部署MySQL，此时应该能正常初始化数据库。

第二步：清理存储卷里的残留损坏数据 #

如果之前部署失败后，存储卷里留下了损坏的InnoDB文件（比如ibdata1、ib_logfile*这类核心文件），哪怕权限正确，MySQL启动时读取这些坏文件还是会崩溃。

操作步骤：

1. 同样用调试容器挂载PVC：

   oc run debug-pod --image=busybox --command -- sleep 3600
   oc rsh debug-pod
   rm -rf /var/lib/mysql/*  # 确认路径正确再执行，避免误删其他数据
   exit
   oc delete pod debug-pod
   

2. 重新部署MySQL，让它从头初始化数据库。

第三步：检查存储类的兼容性 #

有些存储类（比如部分NFS存储）可能不支持MySQL需要的文件系统特性（比如直接IO、文件锁），这也会导致InnoDB初始化失败。

操作步骤：

1. 查看你当前使用的存储类配置：

   oc get storageclass
   oc describe storageclass <你的存储类名称>
   

2. 如果是NFS存储，检查NFS服务器的权限配置，确保支持必要的文件系统特性。如果问题依旧，可以尝试更换存储类（比如本地存储或块存储类型的存储类）。

附：你提供的错误日志 #

=> sourcing http://20-validate-variables.sh ...
=> sourcing http://25-validate-replication-variables.sh ...
=> sourcing http://30-base-config.sh ...
---> 01:27:23 Processing basic MySQL configuration files ...
=> sourcing http://60-replication-config.sh ...
=> sourcing http://70-s2i-config.sh ...
---> 01:27:23 Processing additional arbitrary MySQL configuration provided by s2i ...
=> sourcing 10-mysql57.cnf ...
=> sourcing 40-paas.cnf ...
=> sourcing 50-my-tuning.cnf ...
---> 01:27:23 Initializing database ...
---> 01:27:24 Running /opt/rh/rh-mysql57/root/usr/libexec/mysqld --initialize --datadir=/var/lib/mysql/data --ignore-db-dir=lost+found
---> 01:27:24 Starting MySQL server with disabled networking ...
---> 01:27:24 Waiting for MySQL to start ...
2019-03-30T01:27:24.466755Z 0 [Warning] TIMESTAMP with implicit DEFAULT value is deprecated. Please use --explicit_defaults_for_timestamp server option (see documentation for more details).
2019-03-30T01:27:24.584395Z 0 [Note] /opt/rh/rh-mysql57/root/usr/libexec/mysqld (mysqld 5.7.24) starting as process 40 ...
2019-03-30T01:27:24.774993Z 0 [Note] InnoDB: PUNCH HOLE support available
2019-03-30T01:27:24.775057Z 0 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
2019-03-30T01:27:24.775068Z 0 [Note] InnoDB: Uses event mutexes
2019-03-30T01:27:24.775075Z 0 [Note] InnoDB: GCC builtin __atomic_thread_fence() is used for memory barrier
2019-03-30T01:27:24.775761Z 0 [Note] InnoDB: Compressed tables use zlib 1.2.7
2019-03-30T01:27:24.775809Z 0 [Note] InnoDB: Using Linux native AIO
2019-03-30T01:27:24.777416Z 0 [Note] InnoDB: Number of pools: 1
2019-03-30T01:27:24.778012Z 0 [Note] InnoDB: Not using CPU crc32 instructions
2019-03-30T01:27:24.784562Z 0 [Note] InnoDB: Initializing buffer pool, total size = 256M, instances = 1, chunk size = 128M
2019-03-30T01:27:24.841105Z 0 [Note] InnoDB: Completed initialization of buffer pool
2019-03-30T01:27:24.852695Z 0 [Note] InnoDB: If the mysqld execution user is authorized, page cleaner thread priority can be changed. See the man page of setpriority().
2019-03-30T01:27:25.019343Z 0 [ERROR] InnoDB: Header page consists of zero bytes in datafile: ./ibdata1, Space ID:0, Flags: 0. Please refer to http://dev.mysql.com/doc/refman/5.7/en/innodb-troubleshooting-datadict.html … for how to resolve the issue.
2019-03-30T01:27:25.019422Z 0 [ERROR] InnoDB: Corrupted page [page id: space=0, page number=0] of datafile './ibdata1' could not be found in the doublewrite buffer.
2019-03-30T01:27:25.019473Z 0 [ERROR] InnoDB: Plugin initialization aborted with error Data structure corruption
---> 01:27:25 Waiting for MySQL to start ...
2019-03-30T01:27:25.520693Z 0 [ERROR] Plugin 'InnoDB' init function returned error.
2019-03-30T01:27:25.520749Z 0 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
2019-03-30T01:27:25.520763Z 0 [ERROR] Failed to initialize builtin plugins.
2019-03-30T01:27:25.520772Z 0 [ERROR] Aborting
2019-03-30T01:27:25.521120Z 0 [Note] Binlog end
2019-03-30T01:27:25.521862Z 0 [Note] Shutting down plugin 'CSV'
2019-03-30T01:27:25.530469Z 0 [Note] /opt/rh/rh-mysql57/root/usr/libexec/mysqld: Shutdown complete

内容的提问来源于stack exchange，提问作者user1859366