群晖里面挂载OneDrive到本地目录 + Google Drive（需要先有翻墙服务）

date

Feb 11, 2022

slug

nas-synology-rclone-onedrive-mount2local

status

Published

summary

这样就可以开给Plex之类的来使用5T盘网盘了

Summary

20240201 - fix fusermount3错误 `ln -s /bin/fusermount /bin/fusermount3`

https://imnks.com/8208.html

步骤

在其他地方先获得one drive的正确配置

ssh 登陆到群晖进行命令行配置


ssh admin@10.0.0.18 # 群晖nas ip

export HTTP_PROXY=http://127.0.0.1:18010 
export HTTPS_PROXY=http://127.0.0.1:18010 
export socks5_proxy=socks5://127.0.0.1:18010 
export http_proxy=http://127.0.0.1:15004
export https_proxy=$http_proxy
export socks5_proxy=socks5://127.0.0.1:15004

curl https://rclone.org/install.sh | sudo bash

" # or参照 sudo 安装
# https://github.com/ravem/synology-pcloud-and-rclone
sudo -i
wget https://rclone.org/install.sh
chmod +x install.sh
./install.sh
"

rclone config # to find conf path
vi /var/services/homes/admin/.config/rclone/rclone.conf
# 把配置贴进去即可

# 我希望用root来run，就把rclone相关的配置+cache都放到一个都能访问的地方
cp /var/services/homes/admin/.config/rclone/rclone.conf /volume1/mount/_rclone_config/

# 加载点目录创建
mkdir /volume1/mount/_rclone_od_xyz_e3m1024

# !!! 正式加载（这行也是放到群晖scheduler当中就行）
echo "### rclone od_xyz_e3m1024"; sleep 60s; \
/bin/rclone mount od_xyz_e3m1024: /volume1/mount/_rclone_od_xyz_e3m1024 --copy-links --no-gzip-encoding --no-check-certificate --allow-other --allow-non-empty --umask 000 --vfs-cache-mode writes \
  --vfs-read-chunk-size 32M `# enough to cache `\
  --vfs-read-ahead 128M \
  --attr-timeout 10m \
  --cache-dir /volume1/mount/_rclone_config/db/.cache \
  --dir-cache-time 30m \
  --daemon-wait  2m --daemon --daemon-timeout 30m \
  --log-level=INFO --log-file=/volume1/mount/_rclone_config/rclone_od_xyz_e3m1024.log \
  --config='/volume1/mount/_rclone_config/rclone.conf'; echo "rclone od_xyz_e3m1024 done ###"


# 验证
df -h
rclone lsd od_xyz_e3m1024: --config='/volume1/mount/_rclone_config/rclone.conf'

# 卸载
fusermount -qzu /volume1/mount/_rclone_od_xyz_e3m1024
# 或者
umount -f /volume1/mount/_rclone_od_xyz_e3m1024

在群晖上面配置开机后之后rclone mount

验证

如果是Google Drive - 假设http proxy开在17890端口

用access_token，那就容易，只要加翻墙设置就行了

export DRIVENAME="gd_omgd"
mkdir /volume1/mount/_rclone_$DRIVENAME

export CONF_DIR="/volume1/mount/_rclone_config"
mkdir -p $CONF_DIR/db/.cache

# 在群晖的Scheduler里面设立开机脚本的话，因为考虑此时docker daemon还没有起来（我试了好多次，确定docker起来应该比较后面），前面可以加一个 sleep 60s; 应该可以来确保翻墙的docker已经运行了


echo \
HTTP_PROXY=http://127.0.0.1:18010 \
HTTPS_PROXY=http://127.0.0.1:18010 \
/bin/rclone mount $DRIVENAME: /volume1/mount/_rclone_$DRIVENAME --copy-links --no-gzip-encoding --no-check-certificate --allow-other --allow-non-empty --umask 000 --vfs-cache-mode writes \
  --vfs-read-chunk-size 32M `# enough to cache`\
  --vfs-read-ahead 128M \
  --attr-timeout 10m \
  --cache-dir $CONF_DIR/db/.cache \
  --dir-cache-time 30m \
  --daemon --daemon-timeout 30m \
  --log-level=INFO --log-file=$CONF_DIR/rclone_$DRIVENAME.log \
  --config="$CONF_DIR/rclone.conf";


echo "### rclone $DRIVENAME begin"; \
sleep 60s; \
HTTP_PROXY=http://127.0.0.1:18010 HTTPS_PROXY=http://127.0.0.1:18010 /bin/rclone mount gd_omgd: /volume1/mount/_rclone_gd_omgd --copy-links --no-gzip-encoding --no-check-certificate --allow-other --allow-non-empty --umask 000 --vfs-cache-mode writes --vfs-read-chunk-size 32M --vfs-read-ahead 128M --attr-timeout 10m --cache-dir /volume1/all/rclone/db/.cache --dir-cache-time 30m --daemon --daemon-timeout 30m --log-level=INFO --log-file=/volume1/all/rclone/rclone_gd_omgd.log --config=/volume1/all/rclone/rclone.conf \
echo "rclone $DRIVENAME done ###"

如果是用AutoRclone json 身份文件，用tar先把目录复制过来，再改配置

tar -cvzf 20220212_bak_AutoRclone.tar.gz /root/AutoRclone # use your own    AutoRclone directory

# cp or scp over the file

cp 20220212_bak_AutoRclone.tar.gz /drive/qhwebdav_remote/webdav_remote/

tar -xvzf ~/qh/webdav_remote/20220212_bak_AutoRclone.tar.gz -C ~/qh/all/rclone/

# 这样就实现了从
# Origin_Host:/root/AutoRclone 
# ->
# Dest_Host: ~/qh/all/rclone/AutoRclone
# 的完整迁移