* 2020-03-06 ปิด CI/CD เนื่องจาก กรณี api มีปัญหา restart service ช้ามากจนเกิด timeout หลังจากนั้น git runner จะไม่สามารถรัน job อื่น ๆ ได้อีก

API-Server Installation from source code

0. Prerequisite program and data
	0.1) Install git (>=2.0),
		>> https://tecadmin.net/install-git-2-0-on-centos-rhel-fedora/
   
	0.2) Install go (1.7.x) to ~/go
		0.2.1) Download git for Linux amd64 from https://golang.org/dl/ to ~/go
			$ cd ~
			$ wget https://storage.googleapis.com/golang/go1.7.4.linux-amd64.tar.gz
			$ tar xvfz go1.7.4.linux-amd64.tar.gz
			
		0.2.2) Set go related environment variables in ~/.bash_profile by adding 
			the following lines:
   
			PATH=$HOME/bin:$PATH:$HOME/go/bin
			GOROOT=$HOME/go
			GOPATH=$HOME/go_local
			export PATH GOPATH GOROOT
		
		0.2.3) logout and re login to activate the changes in .bash_profile
   
	0.3) Database connection information. The required information are:
		1. Database host/port
		2. Database name
		3. Database user/password
		Note: database user must be the owner of the database     
		
	0.4) path to server storage on NFS storage server. See 
			https://www.centos.org/docs/5/html/5.1/Deployment_Guide/s2-nfs-config-autofs.html
		For more information about NFS and autofs.	
		For example: /data/thaiwater/thaiwaterdata/serverstore
		Note: user must be able to write to the given NFS storage
		
	0.5) URL to access to the API server cluster
		For example: http://api2.thaiwater30.net:9200
		
	0.6) 256 bits primary encryption key for the cluster in hex string format
		Note: You can use openssl to generate a random key with the following command:
        $ /usr/bin/openssl rand -hex 32		
		
	0.7) IP address/port for a API server to listen to
		For example: http://192.168.12.191:9200
		Note: 0.5) and 0.6) can be the same for the cluster with one server and there is no plan to
		add more servers to the cluster.		
		
	0.8) Unique server name in the cluster
		For example: server1
		
	0.9) Dataimport node name and user/passwor and path to dataimport data on NFS storage server   
        For example: server: 192.168.11.189
					 user: thaiwater30
					 password: abcd
					 path: /home/thaiwater/thaiwaterdata/dataimport
		Note: see dataimport node installation document for information about how to install and
		config dataimport node.	
		   
1. Install api server from source code
	1.1) Create a folder to store source code 
		
		$ mkdir -p ~/go_local/src/haii.or.th/api
		$ cd ~/go_local/src/haii.or.th/api
		
	1.2) Use git clone to initialize api/server,api/util and api/thaiwater30
		
		$ git clone git@git.haii.or.th:cim_cws/api-server.git server
		$ git clone git@git.haii.or.th:cim_cws/api-util.git util
		$ git clone git@git.haii.or.th:cim_cws/api-thaiwater30.git thaiwater30
		
	1.3) Set the code brance of api/server to thaiwater30 , api/util to thaiwater30 
		and thaiwater30 to master branch
		
		$ cd server
		$ git branch thaiwater30
		$ cd ../util
		$ git branch thaiwater30
			
	1.3) Setup symbolic	for Makefile
		
		$ cd thaiwater30
		$ ln -s Makefile.linux Makefile
		
	1.3) Build source code
		
		$ make all
		
		The executables will be saved in ~/go_local/bin folder	  
		
2. Intialize database for server cluster using the sql command files, 
   create a schema and run each file from minimum number prefix to maximum number prefix using psql 
   (except public schema which already existed)
	2.1) api schema
		    Files path:
				schema:	~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/schema
				data: 	~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/data
			
			$ psql
				# CREATE SCHEMA api 
				# \i ~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/schema/0001_trigger.sql
				...
				# \i ~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/schema/6002_user_resetpassword_log.sql
				# \i ~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/data/001_lt_agent_type.sql
				...
				# \i ~/go_local/src/haii.or.th/api/server/tools/installer/initdata/database/0001_api/data/010_agent.sql
		Note: You can use psql ENVIRONMENT variables (https://www.postgresql.org/docs/9.5/static/libpq-envars.html) and the
			following shell script to run all sql files in the folder instead of run each file name with \in
				
				for f in *.sql;
				do
					psql -f "$f"
				done
			
	2.2) public schema (note public schema already existed , no need to run CREATE SCHEMA public command)
			Files path:
				schema:	~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/database/0001_public/schema
				data: 	~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/database/0001_public/data
				
	2.3) dataservice schema 
			schema:	~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/database/0002_dataservice/schema
			data: 	~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/database/0002_dataservice/data
			
3. Create server startup script 
	3.1) Create a script folder
		$ mkdir -p ~/haii-api-server
		
	3.2) Set folder permission to prevent another user to see saved database password and primary encryption key	
		$ chmod og-rwx ~/haii-api-server
		
	3.3) Create a folder to store log files
		$ cd ~/haii-api-server
		$ mkdir logs
		
	3.3) Create haii-api-servce.conf in script folder with following content. Replace database_username, 
		database_password_in_urlencode, host, port and databasename with data acqured from step 0.3 ,
		key from 0.6 and server_name from 0.8 :
		
		export TW30_DB=postgres://database_username:database_password_in_urlencode@host:port/databasename?sslmode=disable
		export TW30_SERVER_AGENT=server_name_from_0_8
		export TW30_KEY=key_from_0_6
		SERVER_NAME=thaiwater30		
		
	3.4) Make a symbolic link from the start up script in source code 
			~/go_local/src/haii.or.th/api/server/tools/script/haii-api-server
		to the script folder
		$ cd ~/haii-api-server
		$ ln -s ../go_local/src/haii.or.th/api/server/tools/script/haii-api-server
		
	3.5) Add user cron job 
			@reboot ~/haii-api-server/haii-api-server start
		to start api server when server starts
		
		$ crontab -e
			@reboot ~/haii-api-server/haii-api-server start
				
	3.6) Add user cron job 
			0 1 * * * ~/go_local/bin/archivelog -logpath=/home/cim/haii-api-server/logs
		to compress old log files	
		
		$ crontab -e
			0 1 * * * ~/go_local/bin/archivelog -logpath=/home/cim/haii-api-server/logs

4. Add cluster specified data
	4.1) Service and groups
		Use ~/go_local/bin/apidbinit and the CSV files in 
			~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/service/service.csv and 
			~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/service/group.csv 
		to set up services and group:
	
			$ ~/go_local/bin/apidbinit -config ~/haii-api-server/haii-api-servce.conf addservice  ~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/service/service.csv
			$ ~/go_local/bin/apidbinit -config ~/haii-api-server/haii-api-servce.conf addgroup  ~/go_local/src/haii.or.th/api/thaiwater30/installer/initdata/service/group.csv
		
	4.2) Web super user and group
		Use apidbinit to add Web super user and group. The default group and super user name is "wheel" and "root".
		Don't forget to replace web_root_password with the real password.
		
		The user "root" is a user that has the permission to acceess to all web menu specified in "group.csv" with category "srv".
		  
			$ ~/go_local/bin/apidbinit -config ~/haii-api-server/haii-api-servce.conf  addroot wheel root web_root_password
		
	4.3) Config dataimport 
		4.3.1) Add remote converter node.
			a) Set upload file path 
				The server support manual dataimport via file upload. We need to set the path to store those files
				in "server.service.dataimport.UploadPathName" setting of api.system_setting table.
				
					$ psql
					#	INSERT INTO api.system_setting SET  
			
		
		... to be written ...
		... wait for download and dataset configuration in CSV format
		
	4.4) Add required system_setting 	
		... to be written .. ... wait for data from programmers about require configs that can not be set from web page ...
		
5. Start server
	7.1) install server executable to script folder and generate swagger document
		$ cd ~/go_local/src/haii.or.th/api/thaiwater30
		$ make pull install install-docs
	  
	7.2) start server
		$ ~/haii-api-server	start
		
	7.3) check server status	
		$ ~/haii-api-server	status
		
	7.4) stop server
		$ ~/haii-api-server	stop
		
6. Install system cronjob to start API server when the server reboot and 
	compress the old log file everyday at 01:00
		$ crontab -e
		  @reboot /home/user_name/haii-api-server/haii-api-server start
		  0 1 * * * /home/user_name/go_local/bin/archivelog -logpath=/home/user_name/haii-api-server/logs
		  (ESC)
		  :wq
		  (ESC)
		  :wq