nginx的ngx_http_upstream_check_module的解析
This project is maintained by wangfakang
对NGX_HTTP_UPSTREAM_CHECK_MODULE的分析：
http {
lua_package_path "/path/to/lua-resty-upstream-healthcheck/lib/?.lua;;";
# sample upstream block:
upstream foo.com {
server 127.0.0.1:12354;
server 127.0.0.1:12355;
server 127.0.0.1:12356 backup;
}
# the size depends on the number of servers in upstream {}:
lua_shared_dict healthcheck 1m;
lua_socket_log_errors off;
init_worker_by_lua '
local hc = require "resty.upstream.healthcheck"
local ok, err = hc.spawn_checker{
shm = "healthcheck", -- defined by "lua_shared_dict"
upstream = "foo.com", -- defined by "upstream"
type = "http",
-- if you put this Lua snippet in separate .lua file,
-- then you should write this instead: http_req = "GET /status HTTP/1.0\r\nHost: foo.com\r\n\r\n",
http_req = "GET /status HTTP/1.0\\r\\nHost: foo.com\\r\\n\\r\\n",
-- raw HTTP request for checking
interval = 2000, -- run the check cycle every 2 sec
timeout = 1000, -- 1 sec is the timeout for network operations
fall = 3, -- # of successive failures before turning a peer down
rise = 2, -- # of successive successes before turning a peer up
valid_statuses = {200, 302}, -- a list valid HTTP status code
concurrency = 10, -- concurrency level for test requests
}
if not ok then
ngx.log(ngx.ERR, "failed to spawn health checker: ", err)
return
end
-- Just call hc.spawn_checker() for more times here if you have
-- more upstream groups to monitor. One call for one upstream group.
-- They can all share the same shm zone without conflicts but they
-- need a bigger shm zone for obvious reasons.
';
server {
...
# status page for all the peers:
location = /status {
access_log off;
allow 127.0.0.1;
deny all;
default_type text/plain;
content_by_lua '
local hc = require "resty.upstream.healthcheck"
ngx.say("Nginx Worker PID: ", ngx.worker.pid())
ngx.print(hc.status_page())
';
}
}
}上面的配置就启用了upstream的健康检查。
注意此模块是agentzh使用nxg_http_lua_upstream_module中的set_peer_down来实现的健康检查.和tengine实现的是不一样的.
local function get_lock(ctx)
local dict = ctx.dict
local key = "l:" .. ctx.upstream
-- the lock is held for the whole interval to prevent multiple
-- worker processes from sending the test request simultaneously.
-- here we substract the lock expiration time by 1ms to prevent
-- a race condition with the next timer event.
-- 利用其过期机制 相当于设置了一个自动过期的锁
local ok, err = dict:add(key, true, ctx.interval - 0.001)
if not ok then
if err == "exists" then
return nil
end
errlog("failed to add key \"", key, "\": ", err)
return nil
end
return true
end
１.如何保证每一个work子进程只执行一次相应的修改?
2.如何知道哪一个peer进行了相应的修改?
在使用中有任何问题，欢迎反馈给我，可以用以下联系方式跟我交流