PHP Multi curl 带不同端口

Question

我正在使用 IOT 系统并尝试改进我们的数据摄取。 目前，我们正在迭代一组具有不同端口的 URL，并一次发送一个 curl 请求。 我想使用 multicurl 来解决这个瓶颈，但我只得到错误作为响应。 这是我的代码（IP 地址已编辑）

$url = 'http://XXX.XXX.XX.XXX/getvar.csv';
$ports = [8101,8102,8103,8104];

foreach ($ports as $port) {
        $worker = curl_init($curl_url);
        curl_setopt_array($worker, [
            CURLOPT_RETURNTRANSFER => true,
            CURLOPT_POST => true,
            CURLOPT_HEADER => 0,
            CURLOPT_TIMEOUT => 8,
            CURLOPT_POSTFIELDS => $content,
            CURLOPT_PORT => $port
        ]);

        curl_multi_add_handle($mh, $worker);
    }
$i=0;
for (; ;) {
    $still_running = null;
    echo "<p>exec multi curl $i</p>";
    do {
        echo "<p>curl multi exec</p>";
        $err = curl_multi_exec($mh, $still_running);
    } while ($err === CURLM_CALL_MULTI_PERFORM);
    if ($err !== CURLM_OK) {
        $description = curl_strerror(curl_errno($mh));
        echo "<p>Error description: $description</p>";
        // handle curl multi error?
    }
    if ($still_running < 1) {
        echo "<p>curl multi finished</p>";
        // all downloads completed
        break;
    }
    // some haven't finished downloading, sleep until more data arrives:
    curl_multi_select($mh, 8);
    $i++;
}

$curl_time_stop = microtime(true);
$curl_duration += $curl_time_stop - $curl_time_start;
echo "CURL finished";
$results = [];
while (false !== ($info = curl_multi_info_read($mh))) {
    if ($info["result"] !== CURLE_OK) {
        echo "Error ".curl_strerror($info["result"]);
        // handle download error?
    }
    $results[curl_getinfo($info["handle"], CURLINFO_EFFECTIVE_URL)] = 
    curl_multi_getcontent($info["handle"]);
    curl_multi_remove_handle($mh, $info["handle"]);
    curl_close($info["handle"]);
}
curl_multi_close($mh);
var_dump($results);

这是我得到的错误/响应

Error Server returned nothing (no headers, no data)
results int(1) int(52) resource(29) of type (curl)

据我所知，错误代码 52 对应于CURLE_GOT_NOTHING 。 有什么我想念的吗？ 还是多卷曲与 curl 使用的协议有区别？

[编辑：添加详细输出] 这是我使用CURLOPT_VERBOSE=>1时的结果

*   Trying 1XX.XX.XX.XX:8101...

* Found bundle for host 1XX.XX.XX.XX: 0x2625a00 [serially]
* Server doesn't support multiplex (yet)
* Hostname 1XX.XX.XX.XX was found in DNS cache
*   Trying 1XX.XX.XX.XX:8101...

*** Connected to 1XX.XX.XX.XX (1XX.XX.XX.XX) port 8101 (#0)
> POST /getvar.csv? HTTP/1.1
Host: 1XX.XX.XX.XX:8101
Accept: */*
Content-Length: 248
Content-Type: application/x-www-form-urlencoded**

* Mark bundle as not supporting multiuse
< HTTP/1.1 200 OK
< Content-Type: text/csv; charset="utf-8"
< Server: embedded HTTPD
< Expires: 1 JAN 2013 00:00:00 GMT
< Last-Modified: 4 MAR 2021 12:10:09 GMT
< Cache-Control: no-cache
< Transfer-Encoding: chunked

* Connection #0 to host 1XX.XX.XX.XX left intact
* Empty reply from server
* Closing connection 25

Answer 1

我发现我的代码存在外部问题，但我在此处发布最终的工作代码以供将来参考。 Multi-curl 不喜欢迭代端口号，所以我不得不将端口烘焙到 curl URL 并迭代 URL 数组。 $content不是 multicurl 工作所必需的，但它是从我正在调用的 URL 中获取数据。

<?php


$lists=[
    "LocalBoardTemp"
];
$url = 'http://XXX.XXX.XXX.XXX:';

$ports = [8101,8102,8103,8104,8105,8106,8107,8108,8109];

foreach ($ports as $port) {
    $curl_urls[] = $url.$port.'/getvar.csv?';
}
$content = '';
foreach ($lists as $list1) {
    $content = $content . "&name=" . $list1;
}

$mh = curl_multi_init();
$fp = fopen(dirname(__FILE__).'/curl_errorlog.txt', 'w');
$i=0;
foreach ($curl_urls as $url) {
    $worker = curl_init($url);
    curl_setopt_array($worker, [
        CURLOPT_RETURNTRANSFER => true,
        CURLOPT_POST => true,
        CURLOPT_HEADER => 0,
        CURLOPT_TIMEOUT => 8,
        CURLOPT_POSTFIELDS => $content,
        CURLOPT_VERBOSE=>1,
        CURLOPT_STDERR => $fp
//        ,
//        CURLOPT_PORT => $port
    ]);
    curl_multi_add_handle($mh, $worker);
}
for (; ;) {
    $still_running = null;
    echo "<p>exec multi curl $i</p>";
    do {
        echo "<p>curl multi exec</p>";
        $err = curl_multi_exec($mh, $still_running);
    } while ($err === CURLM_CALL_MULTI_PERFORM);
    if ($err !== CURLM_OK) {
        $description = curl_strerror(curl_errno($mh));
        echo "<p>Error description: $description</p>";
        // handle curl multi error?
    }
    if ($still_running < 1) {
        echo "<p>curl multi finished</p>";
        // all downloads completed
        break;
    }
    // some haven't finished downloading, sleep until more data arrives:
    curl_multi_select($mh, 8);
    $i++;
}

echo "CURL finished";
$results = [];
$j = 0;
while (false !== ($info = curl_multi_info_read($mh))) {

    if ($info["result"] !== CURLE_OK) {
        echo "Error ".curl_strerror($info["result"]);
        // handle download error?
    }
    echo "<p>results ";
    foreach ($info as $x) {
        var_dump($x);
    }
    echo "</p>";
    $results[curl_getinfo($info["handle"], CURLINFO_EFFECTIVE_URL)] = curl_multi_getcontent($info["handle"]);
    echo "<p>result index $j". curl_multi_getcontent($info["handle"])."</p>";
    curl_multi_remove_handle($mh, $info["handle"]);
    curl_close($info["handle"]);
    $j++;
}
curl_multi_close($mh);
echo "<p>All results dump: ";
var_dump($results);
echo "</p>";
fclose($fp);
echo "finished";

这是由几个教程和堆栈溢出答案拼凑而成的，所以我确信有一种更有效的方法可以做到这一点，但它确实有效。