Overview - Function Compute - Alibaba Cloud Documentation Center

This topic describes how to use a custom runtime to write functions in a Function Compute.

Background

You can use a custom runtime to create a dedicated runtime environment. A custom runtime allows you to:

Use a specified programming language, such as Rust.
Customize a runtime environment of a specified version for a programming language, such as Node.js 16.

Container environment

A custom runtime runs in the following container environment:

Operating system: Debian 9
Architecture: x86_64
User permissions:
- If functions are created at and after 00:00:00 December 1, 2021, the functions must be executed by the root user.
- If functions are created before 00:00:00 December 1, 2021, the functions must be executed by a non-root user.
Directory permissions:
- If functions are created at and after 00:00:00 December 1, 2021, all the directories are writable.
- For the functions that are created before 00:00:00 December 1, 2021, data can be written to only the /tmp directory.
The directory of the code in the container: /code.

Information about the runtime environment

A custom runtime supports the following built-in languages. You can create custom runtimes of the following languages without the need to install third-party interpreters:

Python 3.7.4
Node.js 10.16.2
OpenJDK 1.8.0
Ruby 2.7
PowerShell 7.1.0
Nginx 1.10.3

PHP 7.4.12

Expand to see a list of PHP 7.4.12 built-in extensions.


bcmath	calendar	Core
ctype	curl	date
dom	exif	FFI
fileinfo	filter	ftp
gd	gettext	hash
iconv	imagick	imap
intl	json	libxml
mbstring	mcrypt	memcached
mysqli	mysqlnd	openSSL
pcntl	pcre	PDO
pdo_mysql	pdo_pgsql	pdo_sqlite
pgsql	Phar	posix
protobuf	readline	redis
Reflection	session	shmop
SimpleXML	soap	sockets
sodium	SPL	sqlite3
standard	swoole	sysvmsg
sysvsem	sysvshm	tokenizer
xml	xmlreader	xmlrpc
xmlwriter	xsl	Zend OPcache
zip	zlib	None.

Use a non-built-in programming language

If you want to create a custom runtime by using a non-built-in language of the custom runtime, you must compress the parser or runtime of the language and your code file into a package and deploy the package in Function Compute. For example, if the runtime environment is Node.js 16, you must download the interpreter required for Node.js 16, compress the interpreter and your code file into a package, and then deploy the package in Function Compute. Perform the following steps:

Download the Linux-x64 edition of node to the directory of the code package.

wget http://mirrors.nju.edu.cn/nodejs/v16.14.2/node-v16.14.2-linux-x64.tar.gz -O node-v16.14.2-linux-x64.tar.gz && tar -zxvf node-v16.14.2-linux-x64.tar.gz && rm -rf node-v16.14.2-linux-x64.tar.gz

Configure the startup command by which the custom runtime uses node.

customRuntimeConfig:
  command:
    - /code/node-v16.14.2-linux-x64/bin/node
  args:
    - 'server.js'

# You can also configure environment variables for the function: PATH=/code/node-v16.14.2-linux-x64/bin:/usr/local/bin/apache-maven/bin:/usr/local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/ruby/bin. # After the configuration, you can use node to start the HTTP server.

Install dependencies

For more information about how to install the dependencies of each language in a custom runtime, see the following documentation.

You can also use the layer feature of Function Compute to install dependencies. We recommend that you use an official common layer or build a dependency layer online. For more information, see the following topics: