I'm using Node.js - async & request module to crawl 100+ millions of websites and I keep bumping into errors ESOCKETTIMEDOUT & ETIMEDOUT after few minutes.
It works again after I restart the script. It doesn't seem to be connection limit issue because I can still do resolve4, resolveNs, resolveMx and also curl without delay.
Do you see any issue with the code? or any advice? I'd like to push up the async.queue() concurrency to at least a 1000. Thank you.
var request = require('request'),
async = require('async'),
mysql = require('mysql'),
dns = require('dns'),
url = require('url'),
cheerio = require('cheerio'),
iconv = require('iconv-lite'),
charset = require('charset'),
config = require('./spy.config'),
pool = mysql.createPool(config.db);
iconv.skipDecodeWarning = true;
var queue = async.queue(function (task, cb) {
dns.resolve4('www.' + task.domain, function (err, addresses) {
if (err) {
//
// Do something
//
setImmediate(function () {
cb()
});
} else {
request({
url: 'http://www.' + task.domain,
method: 'GET',
encoding: 'binary',
followRedirect: true,
pool: false,
pool: { maxSockets: 1000 },
timeout: 15000 // 15 sec
}, function (error, response, body) {
//console.info(task);
if (!error) {
// If ok, do something
} else {
// If not ok, do these
console.log(error);
// It keeps erroring here after few minutes, resolve4, resolveNs, resolveMx still work here.
// { [Error: ETIMEDOUT] code: 'ETIMEDOUT' }
// { [Error: ESOCKETTIMEDOUT] code: 'ESOCKETTIMEDOUT' }
var ns = [],
ip = [],
mx = [];
async.parallel([
function (callback) {
// Resolves the domain's name server records
dns.resolveNs(task.domain, function (err, addresses) {
if (!err) {
ns = addresses;
}
callback();
});
}, function (callback) {
// Resolves the domain's IPV4 addresses
dns.resolve4(task.domain, function (err, addresses) {
if (!err) {
ip = addresses;
}
callback();
});
}, function (callback) {
// Resolves the domain's MX records
dns.resolveMx(task.domain, function (err, addresses) {
if (!err) {
addresses.forEach(function (a) {
mx.push(a.exchange);
});
}
callback();
});
}
], function (err) {
if (err) return next(err);
// do something
});
}
setImmediate(function () {
cb()
});
});
}
});
}, 200);
// When the queue is emptied we want to check if we're done
queue.drain = function () {
setImmediate(function () {
checkDone()
});
};
function consoleLog(msg) {
//console.info(msg);
}
function checkDone() {
if (queue.length() == 0) {
setImmediate(function () {
crawlQueue()
});
} else {
console.log("checkDone() not zero");
}
}
function query(sql) {
pool.getConnection(function (err, connection) {
if (!err) {
//console.log(sql);
connection.query(sql, function (err, results) {
connection.release();
});
}
});
}
function crawlQueue() {
pool.getConnection(function (err, connection) {
if (!err) {
var sql = "SELECT * FROM domain last_update < (UNIX_TIMESTAMP() - 2592000) LIMIT 500";
connection.query(sql, function (err, results) {
if (!err) {
if (results.length) {
for (var i = 0, len = results.length; i < len; ++i) {
queue.push({"id": results[i]['id'], "domain": results[i]['domain'] });
}
} else {
process.exit();
}
connection.release();
} else {
connection.release();
setImmediate(function () {
crawlQueue()
});
}
});
} else {
setImmediate(function () {
crawlQueue()
});
}
});
}
setImmediate(function () {
crawlQueue()
});
And the system limits are pretty high.
Limit Soft Limit Hard Limit Units
Max cpu time unlimited unlimited seconds
Max file size unlimited unlimited bytes
Max data size unlimited unlimited bytes
Max stack size 8388608 unlimited bytes
Max core file size 0 unlimited bytes
Max resident set unlimited unlimited bytes
Max processes 257645 257645 processes
Max open files 500000 500000 files
Max locked memory 65536 65536 bytes
Max address space unlimited unlimited bytes
Max file locks unlimited unlimited locks
Max pending signals 257645 257645 signals
Max msgqueue size 819200 819200 bytes
Max nice priority 0 0
Max realtime priority 0 0
Max realtime timeout unlimited unlimited us
sysctl
net.ipv4.ip_local_port_range = 10000 61000
