[DOCKER] How to develop from VScode in a remote destination environment or a remote destination container environment

Introduction

I heard that you can develop in a remote environment using VScode, so I tried it.

Execution environment

[Local PC] ・ Windows 10 pro ・ VScode 1.50.1 ・ OpenSSH_for_Windows_7.7p1, LibreSSL 2.6.5

[Remote destination environment] ・ Ubuntu 18.04 LTS (on GCP) ・ Docker 19.03.13 ・ Python3 series

procedure

1. 1. Remote destination server setup
2. 2. Add package & edit settings on local PC
3. 3. SSH connection from VScode to remote host
4. SSH connection from VScode into the remote container

1. 1. Remote destination server setup

SSH key setting

** Set up for SSH connection to VM instance on GCP ** Refer to [SSH connection to GCP VM instance from local PC].

• The key is generated at this timing. Upload the public key to the server side. The private key will be used later.

Python installation

• To check if you can use the remote environment after connecting from VScode

python

#Check if python3 system is installed
$ sudo python --version
$ sudo python3 --version

#If you don't have python3, do the following
$ sudo apt update
$ sudo apt install -y python3
$ sudo apt install -y python3-pip

Python package installation

python

$ pip3 install numpy pandas beautifulsoup4
#* The purpose of the python package is to check it later, so put in a package that is not on your local PC.

Install docker and create container

** Install docker ** [Systemctl cannot be used on Ubuntu in Docker container] Refer to step 1.

• The docker version is docker = ce = 5: 19.03.14 ~ 3-0 ~ ubuntu-bionic

** Prepare a key pair for SSH connection directly to the container environment ** Key generation is the same as the first step. * The same key as the host can be used, but this time it will be created separately.

** Create container **

python

$ sudo docker container run -it -d -p 20022:22 --name test_con python:3

#Enter the container and install the package
$ sudo docker container exec -it test_con /bin/bash

Commands in the container

apt update 
apt install nano

#SSH key setting
apt install -y openssh-server
mkdir /var/run/sshd
mkdir /root/.ssh/
nano /root/.ssh/authorized_keys
#* Copy the contents of the public key

##python package installation
pip install numpy pandas peewee

#Start SSH server
/usr/sbin/sshd -D

2. 2. Add package & edit settings on local PC

Start VScode and install Remote --SSH

Editing the .ssh/config file on your local PC

Edit the contents of the configuration file referring to the following.

.ssh/config

Host gcp
HostName [GCP external IP]
User [User name for SSH connection]
Port [Port number] * If not stated, it will be port 22.
IdentityFile [path to private key]

Check if SSH connection is possible from the command prompt

command prompt

> ssh gcp

#You will probably be asked to enter a passphrase.

• 1. An error will occur if the access restrictions for the private key are too loose.

python

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@         WARNING: UNPROTECTED PRIVATE KEY FILE!          @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
Permissions for 'C:\\Users\\Path to private key' are too open.

• 2. You cannot connect if the access authority of the remote destination Public key is loose. Must be set to [700] or [744].

python

Permission denied, please try again.

3. 3. SSH connection from VScode to remote host

SSH connection from VScode

A new window will open where you can quickly select Linux in the target host's OS category.

• When connecting for the first time, if you do not select the OS category, it will be Could not establish connection to .... and you will not be able to connect.

** Check if the remote environment is available from bash ** When connected for the first time, TERMINAL is 1.install, so click" + "to add bash. Check if you can use the remote environment properly

python

$ python3

>>> import numpy
>>> from bs4 imoprt BeautfulSoup

** OK if you can do it! ** **

4. SSH connection from VScode into the remote container

Add container connection information to the .ssh/config file on your local PC

Edit the contents of the configuration file referring to the following.

.ssh/config

Host gcp
HostName [GCP external IP]
User [User name for SSH connection]
Port [Port number] * If not stated, it will be port 22.
IdentityFile [path to private key]

Host gcp_container
HostName [GCP external IP]
User [User name to connect to SSH] * Because no user was created in the container this time.[root]
Port [Port number] * Port number on the host side with port mapping (20022 in this case)
IdentityFile [Path to private key] * A key different from the above host

• At this stage, try SSH connection from the command prompt or TeraTerm.

Make an SSH connection from VScode in the same way as ** Step 3 ** and check if the container environment is used properly. After the connection is completed, execute the following from TERMINAL.

python

$ python

>>> import pandas 
>>> imoprt peewee

** OK if you can do it! ** **