多服务器共用数据库时，新增或删除存储，文件列表刷新失效

[ForSourceCodeAnalysis]

Please make sure of the following things

• I have read the documentation.
  我已经阅读了文档。

• I'm sure there are no duplicate issues or discussions.
  我确定没有重复的issue或讨论。

• I'm sure it's due to AList and not something else(such as Network ,Dependencies or Operational).
  我确定是AList的问题，而不是其他原因（例如网络，依赖或操作）。

• I'm sure this issue is not fixed in the latest version.
  我确定这个问题在最新版本中没有被修复。

AList Version / AList 版本

v3.35.0

Driver used / 使用的存储驱动

123pan

Describe the bug / 问题描述

这个问题之前有人提出过，但是当时认为是缓存问题
希望可以在页面上手动清理缓存
这里在描述一次，复现步骤如下：

1. A，B服务共用数据库
2. A中新增存储F，B在存储管理里面刷新列表，可以看到A新增的F，但是在文件列表刷新，无法看到存储F挂载的目录，必须要重启B的alist服务才可以
3. 同理，如果A，B都有存储F，A中删除存储F，B中文件列表也不会删除F的挂载目录

Reproduction / 复现链接

问题已经定位到了，原因如下：

1. 获取存储列表是直接从数据库获取的，所以每次获取存储列表都是最新的
2. 获取文件列表时，使用的是内存中的storagesMap变量，而内存是和服务器绑定的，所以就导致获取文件列表使用的不是最新的
3. 本来通过存储管理界面的Relaod All按钮可以更新内存中的storagesMap，但是更新出现了问题
   func LoadAllStorages(c *gin.Context) { storages, err := db.GetEnabledStorages() if err != nil { log.Errorf("failed get enabled storages: %+v", err) common.ErrorResp(c, err, 500, true) return } conf.StoragesLoaded = false go func(storages []model.Storage) { //这里的更新逻辑是：先卸载存储，再挂载 //而storages是从数据库获取的，这就意味着，A服务删除了存储F，在B服务更新时， //只会更新存在的，而F不在数据库中，也就不会动F，所以更新后，F依然存在storagesMap中 //A中新增存储F时，B中更新会先从storagesMap中查询F的挂载目录，查询不到，直接报错…… //这个逻辑有问题，本身就是要更新storagesMap，却使用了旧的storagesMap查询…… for _, storage := range storages { storageDriver, err := op.GetStorageByMountPath(storage.MountPath) if err != nil { log.Errorf("failed get storage driver: %+v", err) continue } // drop the storage in the driver if err := storageDriver.Drop(context.Background()); err != nil { log.Errorf("failed drop storage: %+v", err) continue } if err := op.LoadStorage(context.Background(), storage); err != nil { log.Errorf("failed get enabled storages: %+v", err) continue } log.Infof("success load storage: [%s], driver: [%s]", storage.MountPath, storage.Driver) } conf.StoragesLoaded = true }(storages) common.SuccessResp(c) }

Config / 配置

无

Logs / 日志

[GIN] 2024/07/21 - 02:43:16 | 200 | 28.722787ms | 192.168.31.40 | POST "/api/admin/storage/load_all"
�[36mINFO�[0m[2024-07-21 02:43:17] success load storage: [/123pan4764], driver: [123Pan]
�[36mINFO�[0m[2024-07-21 02:43:17] success load storage: [/crypt123pan4764], driver: [Crypt]
�[36mINFO�[0m[2024-07-21 02:43:18] success load storage: [/baidubdusername], driver: [BaiduNetdisk]
�[36mINFO�[0m[2024-07-21 02:43:18] success load storage: [/baiduluopo], driver: [BaiduNetdisk]
�[36mINFO�[0m[2024-07-21 02:43:18] success load storage: [/cryptBdLp], driver: [Crypt]
�[36mINFO�[0m[2024-07-21 02:43:19] success load storage: [/123panlu], driver: [123Pan]
�[36mINFO�[0m[2024-07-21 02:43:19] success load storage: [/crypt123panlu], driver: [Crypt]
�[36mINFO�[0m[2024-07-21 02:43:19] success load storage: [/local_us], driver: [Local]
�[31mERRO�[0m[2024-07-21 02:43:19] failed get storage driver: no mount path for an storage is: /ali
github.com/alist-org/alist/v3/internal/op.GetStorageByMountPath
/home/sfy/work/alist/internal/op/storage.go:37
github.com/alist-org/alist/v3/server/handles.LoadAllStorages.func1
/home/sfy/work/alist/server/handles/storage.go:135
runtime.goexit
/usr/local/go/src/runtime/asm_amd64.s:1650

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions]

Hello @, this issue was closed due to inactive more than 52 days. You can reopen or recreate it if you think it should continue. Thank you for your contributions again.