feat: refresh content workflow and verification settings

deploy/docker/ARCHITECTURE.md

@@ -204,10 +204,8 @@ Caddy -> frontend Node server
 当前仓库内已经补了：
 
 - `deploy/scripts/backup/backup-postgres.sh`
-- `deploy/scripts/backup/backup-markdown.sh`
 - `deploy/scripts/backup/backup-media.sh`
 - `deploy/scripts/backup/restore-postgres.sh`
-- `deploy/scripts/backup/restore-markdown.sh`
 - `deploy/scripts/backup/restore-media.sh`
 - `deploy/docker/BACKUP_AND_RECOVERY.md`
 

deploy/docker/BACKUP_AND_RECOVERY.md

@@ -1,13 +1,22 @@
 # 备份与恢复说明
 
-这套博客现在已经有：
+当前站点的内容已经是 **DB-only**：
 
 - PostgreSQL 数据库
-- Markdown 原文内容
+  - 文章结构化字段
+  - 文章 Markdown 原文（`posts.source_markdown`）
+  - 分类 / 标签
+  - 版本历史 / 审计日志 / 订阅数据
+  - 站点配置
 - 媒体文件 / 对象存储
-- 版本历史 / 审计日志 / 订阅数据
 
-所以生产上最重要的不是再多一两个功能，而是**出事后能不能快速恢复**。
+因此生产上最重要的是：
+
+1. **数据库备份**
+2. **媒体资源备份**
+3. 定期做恢复演练
+
+> 不再需要单独备份 `backend/content/posts` 之类的本地 Markdown 目录。
 
 ## 1. 建议的最小备份策略
 
@@ -15,11 +24,7 @@
 - **频率**：每天至少 1 次；高频站点建议每 6~12 小时 1 次
 - **工具**：`pg_dump --format=custom`
 - **脚本**：`deploy/scripts/backup/backup-postgres.sh`
-
-### Markdown 原文
-- **频率**：每次发布后 + 每天定时 1 次
-- **脚本**：`deploy/scripts/backup/backup-markdown.sh`
-- **原因**：Markdown 是内容源，恢复速度最快
+- **说明**：文章内容原文已经跟随数据库一起备份
 
 ### 媒体文件
 - 如果是本地目录：打包归档
@@ -35,9 +40,6 @@
 # 单独备份数据库
 DATABASE_URL=postgres://... ./deploy/scripts/backup/backup-postgres.sh
 
-# 单独备份 Markdown
-MARKDOWN_SOURCE_DIR=./backend/content/posts ./deploy/scripts/backup/backup-markdown.sh
-
 # 单独备份媒体（本地目录）
 MEDIA_SOURCE_DIR=./uploads ./deploy/scripts/backup/backup-media.sh
 
@@ -53,12 +55,6 @@ MEDIA_S3_SOURCE=s3://bucket-name ./deploy/scripts/backup/backup-media.sh
 DATABASE_URL=postgres://... ./deploy/scripts/backup/restore-postgres.sh ./backups/postgres/latest.dump
 ```
 
-### 恢复 Markdown
-
-```bash
-MARKDOWN_TARGET_DIR=./backend/content/posts ./deploy/scripts/backup/restore-markdown.sh ./backups/markdown/latest.tar.gz
-```
-
 ### 恢复媒体
 
 ```bash
@@ -75,9 +71,6 @@ MEDIA_S3_TARGET=s3://bucket-name ./deploy/scripts/backup/restore-media.sh ./back
 # 每天 03:10 备份 PostgreSQL
 10 3 * * * cd /opt/termi-astro && DATABASE_URL=postgres://... ./deploy/scripts/backup/backup-postgres.sh >> /var/log/termi-backup.log 2>&1
 
-# 每天 03:25 备份 Markdown
-25 3 * * * cd /opt/termi-astro && MARKDOWN_SOURCE_DIR=./backend/content/posts ./deploy/scripts/backup/backup-markdown.sh >> /var/log/termi-backup.log 2>&1
-
 # 每天 03:40 备份媒体
 40 3 * * * cd /opt/termi-astro && MEDIA_S3_SOURCE=s3://bucket-name ./deploy/scripts/backup/backup-media.sh >> /var/log/termi-backup.log 2>&1
 
@@ -88,7 +81,7 @@ MEDIA_S3_TARGET=s3://bucket-name ./deploy/scripts/backup/restore-media.sh ./back
 40 4 * * * cd /opt/termi-astro && OFFSITE_TARGET=/mnt/offsite/termi-astro-backups ./deploy/scripts/backup/sync-backups-offsite.sh >> /var/log/termi-backup.log 2>&1
 ```
 
-## 5. 建议你们再加一层异地备份
+## 5. 建议再加一层异地备份
 
 仅仅把备份留在同一台服务器上不够。
 
@@ -101,9 +94,8 @@ MEDIA_S3_TARGET=s3://bucket-name ./deploy/scripts/backup/restore-media.sh ./back
 建议每个月至少做 1 次演练：
 
 1. 用最新数据库备份恢复到临时环境
-2. 用 Markdown 备份恢复内容目录
-3. 用媒体备份恢复对象
-4. 校验：
+2. 用媒体备份恢复对象
+3. 校验：
    - 首页可打开
    - 文章详情可打开
    - 图片可访问
@@ -115,7 +107,6 @@ MEDIA_S3_TARGET=s3://bucket-name ./deploy/scripts/backup/restore-media.sh ./back
 ```bash
 DATABASE_URL=postgres://... \
 POSTGRES_BACKUP=./backups/postgres/latest.dump \
-MARKDOWN_BACKUP=./backups/markdown/latest.tar.gz \
 MEDIA_BACKUP=./backups/media/latest.tar.gz \
 ./deploy/scripts/backup/verify-restore.sh
 ```
@@ -125,17 +116,16 @@ MEDIA_BACKUP=./backups/media/latest.tar.gz \
 发生事故时建议按这个顺序：
 
 1. 恢复数据库
-2. 恢复 Markdown 原文
-3. 恢复媒体资源
-4. 启动 backend / frontend / admin
-5. 进入后台检查：
+2. 恢复媒体资源
+3. 启动 backend / frontend / admin
+4. 进入后台检查：
    - 审计日志
    - 文章版本历史
    - 订阅目标与最近投递
 
 ## 8. 说明
 
-这些脚本是**仓库内参考实现**，没有在你们生产机上自动执行。
+这些脚本是**仓库内参考实现**，没有在生产机上自动执行。
 正式上线前请按你们实际目录、R2/S3 桶、数据库连接串、cron 规范再过一遍。
 
 另外仓库里已经提供：

deploy/scripts/backup/backup-all.sh

@@ -4,7 +4,6 @@ set -euo pipefail
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 
 "${SCRIPT_DIR}/backup-postgres.sh"
-"${SCRIPT_DIR}/backup-markdown.sh"
 "${SCRIPT_DIR}/backup-media.sh"
 
 echo "All backup jobs finished successfully."

deploy/scripts/backup/backup-markdown.sh

@@ -1,20 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-SOURCE_DIR="${MARKDOWN_SOURCE_DIR:-./backend/content/posts}"
-BACKUP_DIR="${BACKUP_DIR:-./backups/markdown}"
-RETENTION_DAYS="${RETENTION_DAYS:-30}"
-TIMESTAMP="$(date -u +%Y%m%dT%H%M%SZ)"
-FILE_PATH="${BACKUP_DIR}/markdown-${TIMESTAMP}.tar.gz"
-
-if [[ ! -d "${SOURCE_DIR}" ]]; then
-  echo "Markdown source directory not found: ${SOURCE_DIR}" >&2
-  exit 1
-fi
-
-mkdir -p "${BACKUP_DIR}"
-tar -czf "${FILE_PATH}" -C "${SOURCE_DIR}" .
-ln -sfn "$(basename "${FILE_PATH}")" "${BACKUP_DIR}/latest.tar.gz"
-find "${BACKUP_DIR}" -type f -name 'markdown-*.tar.gz' -mtime +"${RETENTION_DAYS}" -delete
-
-echo "Markdown backup written to ${FILE_PATH}"

deploy/scripts/backup/prune-backups.sh

@@ -3,7 +3,6 @@ set -euo pipefail
 
 BACKUP_ROOT="${BACKUP_ROOT:-./backups}"
 POSTGRES_RETENTION_DAYS="${POSTGRES_RETENTION_DAYS:-14}"
-MARKDOWN_RETENTION_DAYS="${MARKDOWN_RETENTION_DAYS:-30}"
 MEDIA_RETENTION_DAYS="${MEDIA_RETENTION_DAYS:-14}"
 DRY_RUN="${DRY_RUN:-false}"
 
@@ -42,7 +41,6 @@ prune_dirs() {
 }
 
 prune "${BACKUP_ROOT}/postgres" 'postgres-*.dump' "${POSTGRES_RETENTION_DAYS}"
-prune "${BACKUP_ROOT}/markdown" 'markdown-*.tar.gz' "${MARKDOWN_RETENTION_DAYS}"
 prune "${BACKUP_ROOT}/media" 'media-*.tar.gz' "${MEDIA_RETENTION_DAYS}"
 prune_dirs "${BACKUP_ROOT}/media" 'media-*' "${MEDIA_RETENTION_DAYS}"
 

deploy/scripts/backup/restore-markdown.sh

@@ -1,20 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-if [[ $# -lt 1 ]]; then
-  echo "Usage: $0 <backup-file.tar.gz>" >&2
-  exit 1
-fi
-
-TARGET_DIR="${MARKDOWN_TARGET_DIR:-./backend/content/posts}"
-BACKUP_FILE="$1"
-
-if [[ ! -f "${BACKUP_FILE}" ]]; then
-  echo "Backup file not found: ${BACKUP_FILE}" >&2
-  exit 1
-fi
-
-mkdir -p "${TARGET_DIR}"
-rm -rf "${TARGET_DIR}"/*
-tar -xzf "${BACKUP_FILE}" -C "${TARGET_DIR}"
-echo "Markdown restore completed into ${TARGET_DIR}"

deploy/scripts/backup/verify-restore.sh

@@ -3,15 +3,12 @@ set -euo pipefail
 
 : "${DATABASE_URL:?DATABASE_URL is required}"
 : "${POSTGRES_BACKUP:?POSTGRES_BACKUP is required}"
-: "${MARKDOWN_BACKUP:?MARKDOWN_BACKUP is required}"
 : "${MEDIA_BACKUP:?MEDIA_BACKUP is required}"
 
 POSTGRES_RESTORE_CMD="${POSTGRES_RESTORE_CMD:-./deploy/scripts/backup/restore-postgres.sh}"
-MARKDOWN_RESTORE_CMD="${MARKDOWN_RESTORE_CMD:-./deploy/scripts/backup/restore-markdown.sh}"
 MEDIA_RESTORE_CMD="${MEDIA_RESTORE_CMD:-./deploy/scripts/backup/restore-media.sh}"
 
 "${POSTGRES_RESTORE_CMD}" "${POSTGRES_BACKUP}"
-"${MARKDOWN_RESTORE_CMD}" "${MARKDOWN_BACKUP}"
 "${MEDIA_RESTORE_CMD}" "${MEDIA_BACKUP}"
 
 echo "Restore rehearsal completed. Please verify homepage, article detail, media assets, admin login, revisions, audit logs, and subscriptions manually."