Updating your upstream list and run lua scripts without reloading Nginx.
Slardar is a HTTP load balancer based on Nginx, lua-nginx-module and stream-lua-nginx-module, by which you can update your upstream list and run lua scripts without reloading Nginx.
This bundle is maintained by UPYUN(又拍云) Inc.
Because most of the nginx modules are developed by the bundle maintainers, it can ensure that all these modules are played well together.
The bundled software components are copyrighted by the respective copyright holders.
1. Clone the repository
git clone https://github.com/upyun/slardar.git
2. Set installation directory (optional)
By default, Slardar will be installed to /usr/local/slardar
, and you should ensure that you have write permission to the directory.
If you want to change to another location, you should export the PREFIX
environment variable to the path you want to install.
export PREFIX=/path/to/your/dir
3. Configure
cd slardar
make configure
4. Build and Install
make
make install
5. Run
/usr/local/slardar/nginx/sbin/nginx
or you have changed installation directory in step 2:
$PREFIX/nginx/sbin/nginx
1. Clone the repository
git clone https://github.com/upyun/slardar.git
2. Build docker image
cd slardar
docker build -t slardar .
3. Run
docker run -d --net=host --name slardar slardar
Contiguration file is in lua
format and located at /usr/local/slardar/nginx/app/etc/config.lua
or $PREFIX/nginx/app/etc/config.lua
if you changed your installation location.
Example configuration and the comments are listed as follows.
local _M = {}
_M.global = {
-- checkups send heartbeats to backend servers every 5s.
checkup_timer_interval = 5,
-- checkups timer key will expire in every 60s.
-- In most cases, you don't need to change this value.
checkup_timer_overtime = 60,
-- checkups will sent heartbeat to servers by default.
default_heartbeat_enable = true,
-- create upstream syncer for each worker.
-- If set to false, dynamic upstream will not work properly.
-- This switch is used for compatibility purpose only in checkups,
-- don't change this in slardar.
checkup_shd_sync_enable = true,
-- sync upstream list from shared memory every 1s
shd_config_timer_interval = 1,
}
_M.consul = {
-- connect to consul will timeout in 5s.
timeout = 5,
-- disable checkups heartbeat to consul.
enable = false,
-- consul k/v prefix.
-- Slardar will read upstream list from config/slardar/upstreams.
-- For more information, please refer to 'Consul configuration'.
config_key_prefix = "config/slardar/",
-- positive cache ttl(in seconds) for dynamic configurations from consul.
config_positive_ttl = 10,
-- negative cache ttl(in seconds) for dynamic configurations from consul.
config_negative_ttl = 5,
-- cache dynamic configurations from consul.
config_cache_enable = true,
cluster = {
{
servers = {
-- change these to your own consul http addresses
{ host = "10.0.5.108", port = 8500 },
{ host = "10.0.5.109", port = 8500 },
},
},
},
}
return _M
Slardar will read persisted configurations, upstream list and lua code from consul on startup. Consul configuration can be customized by setting k/v with the prefix config_key_prefix
(e.g.config/slardar/
) configured in config.lua
. You should ensure that all values behind config_key_prefix
are in valid json
format.
An example Consul keys and their corresponding values are listed as follows,
consul k/v key | value |
---|---|
lua/modules.abc | local f = {version=10} return f |
lua/script.test | local f = require("modules.abc") print(f.version) |
upstreams/node-dev.example | {"enable": true, "servers": [{"host": "127.0.0.1","port": 8001,"weight": 1,"max_fails": 6,"fail_timeout": 30}]} |
myargs | {"arg0": 0,"arg1": 1} |
For the above example, Slardar will load modules.abc
, script.test
as lua code and node-dev.example
as upstream on startup.
You can set "enable": false
(default is true
) in your upstream configuration to disable periodical heartbeats to servers by checkups.
When Slardar is running, you can use slardar.myargs.arg0
to get arg0
and slardar.myargs.arg1
to get arg1
. The config will be cached for config_positive_ttl
seconds. That is to say, when you change the value of myargs
in consul, it will take effect in config_positive_ttl
seconds.
Differs to configurations like myargs
, keys behind lua
and upstreams
will not be cached and you can only update them by Slardar's HTTP interfaces.
If you don't need any preload scripts or upstreams, just leave nothing behind config_key_prefix
or an empty value.
Slardar is 100% compatible with nginx, so you can change nginx configuration files in the same way you do for Nginx.
Configuration files for Nginx are located at /usr/local/slardar/nginx/conf
or $PREFIX/nginx/conf
if you changed your installation location.
GET 127.0.0.1:1995/status
Slardar will return its status in json format.
{
-- checkups heartbeat timer is alive.
"checkup_timer_alive": true,
-- last heartbeat time
"last_check_time": "2016-08-12 13:09:40",
-- slardar version
"slardar_version": "1.0.0",
-- start or reload time.
"start_time": "2016-08-12 13:09:40",
-- lua config file version, you can set 'conf_hash = "your-version"' in your lua config file.
"conf_hash": null,
-- every time you update upstream, this value will increase.
"shd_config_version": 0,
-- status for consul cluster
"cls:consul": [
[
{
"server": "consul:10.0.5.108:8500",
"weight": 1,
"status": "unchecked"
}
]
],
-- status for node-dev cluster
"cls:node-dev": [
[
{
"server": "node-dev:10.0.5.108:8001",
"weight": 1,
"fail_timeout": 30,
"status": "ok",
"max_fails": 6
}
]
]
}
GET 127.0.0.1:1995/lua
Slardar will return loaded lua scripts and modules in json format.
{
-- every time you update lua scripts, this value will increase.
"version": 0,
"modules": [
{
-- module load time
"time": "2016-08-12 13:09:40",
-- md5 value of module file
"version": "aed4a968ef14f8db732e3602c34dc37a",
-- module name
"name": "modules.test"
},
{
"time": "2016-08-12 13:09:40",
"version": "302f9bf40fcd3734cab120b97f18edf3",
"name": "script.test"
}
]
}
POST 127.0.0.1:1995/upstream/name
The request body is your new upstream list in json format. For example,
curl 127.0.0.1:1995/upstream/node-dev.example.com -d \
{"servers":[{"host":"192.168.1.1", "port": 8080}, {"host":"192.168.1.2", "port": 8080}]}
The example above will add two servers into upstream named node-dev.example.com
.
DELETE 127.0.0.1:1995/upstream/name
For example,
curl -XDELETE 127.0.0.1:1995/upstream/node-dev.example.com
The example above will delete the upstream named node-dev.example.com
.
POST 127.0.0.1:1995/lua/scripts.name
or post a lua module,
POST 127.0.0.1:1995/lua/modules.name
The request body is the lua code of your script or module. For example,
curl 127.0.0.1:1995/lua/scripts.test -d 'return slardar.exit(errno.EXIT_TRY_CODE)'
curl 127.0.0.1:1995/lua/modules.test -d 'local f = {version=10} return f'
PUT 127.0.0.1:1995/lua/scripts.name
or load a lua module,
PUT 127.0.0.1:1995/lua/modules.name
Before loading lua, you must post the lua script or module to Slardar.
For example,
curl -XPUT 127.0.0.1:1995/lua/scripts.test
curl -XPUT 127.0.0.1:1995/lua/modules.test
Get from upstream which does not exist will result in 502.
$ curl 127.0.0.1:8080/ -H "Host: node-dev.example.com"
<html>
<head><title>502 Bad Gateway</title></head>
<body bgcolor="white">
<center><h1>502 Bad Gateway</h1></center>
<hr><center>slardar/1.0</center>
</body>
</html>
Add one server to node-dev.example.com
$ curl 127.0.0.1:1995/upstream/node-dev.example.com -d '{"servers":[{"host":"127.0.0.1", "port": 4000}]}'
{"status":200}
Now, we can get the correct result.
$ curl 127.0.0.1:8080/ -H "Host: node-dev.example.com"
hello world
Load a lua script
$ curl 127.0.0.1:1995/lua/script.node-dev.example.com -d 'if ngx.req.get_method() == "DELETE" then return ngx.exit(403) end'
"ok"
$ curl -XPUT 127.0.0.1:1995/lua/script.node-dev.example.com
The script is taking effect.
$ curl -XDELETE 127.0.0.1:8080/ -H "Host: node-dev.example.com"
<html>
<head><title>403 Forbidden</title></head>
<body bgcolor="white">
<center><h1>403 Forbidden</h1></center>
<hr><center>slardar/1.0</center>
</body>
</html>
This bundle contains only tests for Slardar, the bundled components are tested in their own project.
You can run tests for Slardar by the following commands.
docker run -d --net=host consul agent -dev -bind=127.0.0.1
make dev
make test
You're very welcome to report issues on GitHub.
PRs are more than welcome. Just fork, create a feature branch, and open a PR. We love PRs. :)
The bundle itself is licensed under the 2-clause BSD license.
Copyright (c) 2016, UPYUN(又拍云) Inc.
This module is licensed under the terms of the BSD license.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
- Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
- Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.