Add support for saving video files to object storage (#4290)

* Add support for saving video files to object storage

* Add support for custom url generation on s3 stored files

Uses two config keys to support url generation that doesn't directly go
to (compatible s3). Can be used to generate urls to any cache server or
CDN.

* Upload files to s3 concurrently and delete originals afterwards

* Only publish after move to object storage is complete

* Use base url instead of url template

* Fix mistyped config field

* Add rudenmentary way to download before transcode

* Implement Chocobozzz suggestions

https://github.com/Chocobozzz/PeerTube/pull/4290#issuecomment-891670478

The remarks in question:
    Try to use objectStorage prefix instead of s3 prefix for your function/variables/config names
    Prefer to use a tree for the config: s3.streaming_playlists_bucket -> object_storage.streaming_playlists.bucket
    Use uppercase for config: S3.STREAMING_PLAYLISTS_BUCKETINFO.bucket -> OBJECT_STORAGE.STREAMING_PLAYLISTS.BUCKET (maybe BUCKET_NAME instead of BUCKET)
    I suggest to rename moveJobsRunning to pendingMovingJobs (or better, create a dedicated videoJobInfo table with a pendingMove & videoId columns so we could also use this table to track pending transcoding jobs)
    https://github.com/Chocobozzz/PeerTube/pull/4290/files#diff-3e26d41ca4bda1de8e1747af70ca2af642abcc1e9e0bfb94239ff2165acfbde5R19 uses a string instead of an integer
    I think we should store the origin object storage URL in fileUrl, without base_url injection. Instead, inject the base_url at "runtime" so admins can easily change this configuration without running a script to update DB URLs

* Import correct function

* Support multipart upload

* Remove import of node 15.0 module stream/promises

* Extend maximum upload job length

Using the same value as for redundancy downloading seems logical

* Use dynamic part size for really large uploads

Also adds very small part size for local testing

* Fix decreasePendingMove query

* Resolve various PR comments

* Move to object storage after optimize

* Make upload size configurable and increase default

* Prune webtorrent files that are stored in object storage

* Move files after transcoding jobs

* Fix federation

* Add video path manager

* Support move to external storage job in client

* Fix live object storage tests

Co-authored-by: Chocobozzz <me@florianbigard.com>

6 files changed, 114 insertions, 33 deletions

diff --git a/server/initializers/checker-after-init.ts b/server/initializers/checker-after-init.ts
index 911734fa0..09f587274 100644
--- a/server/initializers/checker-after-init.ts
+++ b/server/initializers/checker-after-init.ts
@@ -153,6 +153,29 @@ function checkConfig () {
     }
   }
 
+  // Object storage
+  if (CONFIG.OBJECT_STORAGE.ENABLED === true) {
+
+    if (!CONFIG.OBJECT_STORAGE.VIDEOS.BUCKET_NAME) {
+      return 'videos_bucket should be set when object storage support is enabled.'
+    }
+
+    if (!CONFIG.OBJECT_STORAGE.STREAMING_PLAYLISTS.BUCKET_NAME) {
+      return 'streaming_playlists_bucket should be set when object storage support is enabled.'
+    }
+
+    if (
+      CONFIG.OBJECT_STORAGE.VIDEOS.BUCKET_NAME === CONFIG.OBJECT_STORAGE.STREAMING_PLAYLISTS.BUCKET_NAME &&
+      CONFIG.OBJECT_STORAGE.VIDEOS.PREFIX === CONFIG.OBJECT_STORAGE.STREAMING_PLAYLISTS.PREFIX
+    ) {
+      if (CONFIG.OBJECT_STORAGE.VIDEOS.PREFIX === '') {
+        return 'Object storage bucket prefixes should be set when the same bucket is used for both types of video.'
+      } else {
+        return 'Object storage bucket prefixes should be set to different values when the same bucket is used for both types of video.'
+      }
+    }
+  }
+
   return null
 }
 
diff --git a/server/initializers/config.ts b/server/initializers/config.ts
index 30a9823b9..0e684eef8 100644
--- a/server/initializers/config.ts
+++ b/server/initializers/config.ts
@@ -73,6 +73,26 @@ const CONFIG = {
     PLUGINS_DIR: buildPath(config.get<string>('storage.plugins')),
     CLIENT_OVERRIDES_DIR: buildPath(config.get<string>('storage.client_overrides'))
   },
+  OBJECT_STORAGE: {
+    ENABLED: config.get<boolean>('object_storage.enabled'),
+    MAX_UPLOAD_PART: bytes.parse(config.get<string>('object_storage.max_upload_part')),
+    ENDPOINT: config.get<string>('object_storage.endpoint'),
+    REGION: config.get<string>('object_storage.region'),
+    CREDENTIALS: {
+      ACCESS_KEY_ID: config.get<string>('object_storage.credentials.access_key_id'),
+      SECRET_ACCESS_KEY: config.get<string>('object_storage.credentials.secret_access_key')
+    },
+    VIDEOS: {
+      BUCKET_NAME: config.get<string>('object_storage.videos.bucket_name'),
+      PREFIX: config.get<string>('object_storage.videos.prefix'),
+      BASE_URL: config.get<string>('object_storage.videos.base_url')
+    },
+    STREAMING_PLAYLISTS: {
+      BUCKET_NAME: config.get<string>('object_storage.streaming_playlists.bucket_name'),
+      PREFIX: config.get<string>('object_storage.streaming_playlists.prefix'),
+      BASE_URL: config.get<string>('object_storage.streaming_playlists.base_url')
+    }
+  },
   WEBSERVER: {
     SCHEME: config.get<boolean>('webserver.https') === true ? 'https' : 'http',
     WS: config.get<boolean>('webserver.https') === true ? 'wss' : 'ws',
diff --git a/server/initializers/constants.ts b/server/initializers/constants.ts
index 5f121d9a4..8a1526ae8 100644
--- a/server/initializers/constants.ts
+++ b/server/initializers/constants.ts
@@ -24,7 +24,7 @@ import { CONFIG, registerConfigChangedHandler } from './config'
 
 // ---------------------------------------------------------------------------
 
-const LAST_MIGRATION_VERSION = 655
+const LAST_MIGRATION_VERSION = 660
 
 // ---------------------------------------------------------------------------
 
@@ -147,7 +147,8 @@ const JOB_ATTEMPTS: { [id in JobType]: number } = {
   'videos-views': 1,
   'activitypub-refresher': 1,
   'video-redundancy': 1,
-  'video-live-ending': 1
+  'video-live-ending': 1,
+  'move-to-object-storage': 3
 }
 // Excluded keys are jobs that can be configured by admins
 const JOB_CONCURRENCY: { [id in Exclude<JobType, 'video-transcoding' | 'video-import'>]: number } = {
@@ -162,7 +163,8 @@ const JOB_CONCURRENCY: { [id in Exclude<JobType, 'video-transcoding' | 'video-im
   'videos-views': 1,
   'activitypub-refresher': 1,
   'video-redundancy': 1,
-  'video-live-ending': 10
+  'video-live-ending': 10,
+  'move-to-object-storage': 1
 }
 const JOB_TTL: { [id in JobType]: number } = {
   'activitypub-http-broadcast': 60000 * 10, // 10 minutes
@@ -178,7 +180,8 @@ const JOB_TTL: { [id in JobType]: number } = {
   'videos-views': undefined, // Unlimited
   'activitypub-refresher': 60000 * 10, // 10 minutes
   'video-redundancy': 1000 * 3600 * 3, // 3 hours
-  'video-live-ending': 1000 * 60 * 10 // 10 minutes
+  'video-live-ending': 1000 * 60 * 10, // 10 minutes
+  'move-to-object-storage': 1000 * 60 * 60 * 3 // 3 hours
 }
 const REPEAT_JOBS: { [ id: string ]: EveryRepeatOptions | CronRepeatOptions } = {
   'videos-views': {
@@ -412,7 +415,8 @@ const VIDEO_STATES: { [ id in VideoState ]: string } = {
   [VideoState.TO_TRANSCODE]: 'To transcode',
   [VideoState.TO_IMPORT]: 'To import',
   [VideoState.WAITING_FOR_LIVE]: 'Waiting for livestream',
-  [VideoState.LIVE_ENDED]: 'Livestream ended'
+  [VideoState.LIVE_ENDED]: 'Livestream ended',
+  [VideoState.TO_MOVE_TO_EXTERNAL_STORAGE]: 'To move to an external storage'
 }
 
 const VIDEO_IMPORT_STATES: { [ id in VideoImportState ]: string } = {
diff --git a/server/initializers/database.ts b/server/initializers/database.ts
index 38e7a76d0..0e690f6ae 100644
--- a/server/initializers/database.ts
+++ b/server/initializers/database.ts
@@ -45,6 +45,7 @@ import { VideoTagModel } from '../models/video/video-tag'
 import { VideoViewModel } from '../models/video/video-view'
 import { CONFIG } from './config'
 import { ActorCustomPageModel } from '@server/models/account/actor-custom-page'
+import { VideoJobInfoModel } from '@server/models/video/video-job-info'
 
 require('pg').defaults.parseInt8 = true // Avoid BIGINT to be converted to string
 
@@ -143,7 +144,8 @@ async function initDatabaseModels (silent: boolean) {
     TrackerModel,
     VideoTrackerModel,
     PluginModel,
-    ActorCustomPageModel
+    ActorCustomPageModel,
+    VideoJobInfoModel
   ])
 
   // Check extensions exist in the database
diff --git a/server/initializers/migrations/0065-video-file-size.ts b/server/initializers/migrations/0065-video-file-size.ts
index 1aeb27f2d..ac952a98c 100644
--- a/server/initializers/migrations/0065-video-file-size.ts
+++ b/server/initializers/migrations/0065-video-file-size.ts
@@ -1,7 +1,4 @@
 import * as Sequelize from 'sequelize'
-import { stat } from 'fs-extra'
-import { VideoModel } from '../../models/video/video'
-import { getVideoFilePath } from '@server/lib/video-paths'
 
 function up (utils: {
   transaction: Sequelize.Transaction
@@ -9,30 +6,7 @@ function up (utils: {
   sequelize: Sequelize.Sequelize
   db: any
 }): Promise<void> {
-  return utils.db.Video.listOwnedAndPopulateAuthorAndTags()
-    .then((videos: VideoModel[]) => {
-      const tasks: Promise<any>[] = []
-
-      videos.forEach(video => {
-        video.VideoFiles.forEach(videoFile => {
-          const p = new Promise((res, rej) => {
-            stat(getVideoFilePath(video, videoFile), (err, stats) => {
-              if (err) return rej(err)
-
-              videoFile.size = stats.size
-              videoFile.save().then(res).catch(rej)
-            })
-          })
-
-          tasks.push(p)
-        })
-      })
-
-      return tasks
-    })
-    .then((tasks: Promise<any>[]) => {
-      return Promise.all(tasks)
-    })
+  throw new Error('Removed, please upgrade from a previous version first.')
 }
 
 function down (options) {
diff --git a/server/initializers/migrations/0660-object-storage.ts b/server/initializers/migrations/0660-object-storage.ts
new file mode 100644
index 000000000..c815c71c6
--- /dev/null
+++ b/server/initializers/migrations/0660-object-storage.ts
@@ -0,0 +1,58 @@
+import * as Sequelize from 'sequelize'
+import { VideoStorage } from '@shared/models'
+
+async function up (utils: {
+  transaction: Sequelize.Transaction
+  queryInterface: Sequelize.QueryInterface
+  sequelize: Sequelize.Sequelize
+  db: any
+}): Promise<void> {
+  {
+    const query = `
+    CREATE TABLE IF NOT EXISTS "videoJobInfo" (
+      "id" serial,
+      "pendingMove" INTEGER NOT NULL,
+      "pendingTranscode" INTEGER NOT NULL,
+      "videoId" serial UNIQUE NOT NULL REFERENCES "video" ("id") ON DELETE CASCADE ON UPDATE CASCADE,
+      "createdAt" timestamp WITH time zone NOT NULL,
+      "updatedAt" timestamp WITH time zone NOT NULL,
+      PRIMARY KEY ("id")
+    );
+    `
+
+    await utils.sequelize.query(query)
+  }
+
+  {
+    await utils.queryInterface.addColumn('videoFile', 'storage', { type: Sequelize.INTEGER, allowNull: true })
+  }
+  {
+    await utils.sequelize.query(
+      `UPDATE "videoFile" SET "storage" = ${VideoStorage.FILE_SYSTEM}`
+    )
+  }
+  {
+    await utils.queryInterface.changeColumn('videoFile', 'storage', { type: Sequelize.INTEGER, allowNull: false })
+  }
+
+  {
+    await utils.queryInterface.addColumn('videoStreamingPlaylist', 'storage', { type: Sequelize.INTEGER, allowNull: true })
+  }
+  {
+    await utils.sequelize.query(
+      `UPDATE "videoStreamingPlaylist" SET "storage" = ${VideoStorage.FILE_SYSTEM}`
+    )
+  }
+  {
+    await utils.queryInterface.changeColumn('videoStreamingPlaylist', 'storage', { type: Sequelize.INTEGER, allowNull: false })
+  }
+}
+
+function down (options) {
+  throw new Error('Not implemented.')
+}
+
+export {
+  up,
+  down
+}