CentOS VirtualBox对http请求的第二个延迟 [英] CentOS VirtualBox 5 second latency on http requests

问题描述

我在我的OSX 10.8.2机器上运行centos 6.3在VirtualBox 4.2.1内，并遇到一个延迟问题，我不理解。基本上，每个http请求到任何地方有一个额外的5000ms延迟。 Ping没有额外延迟。

  [vagrant @ localhost〜] $ curl -o / dev / null https://google.com -w 
％总％接收％Xferd平均速度时间时间时间当前
载入上传总剩余速度
 110 220 110 220 0 0 42 0 0:00:05 0:00:05  - ： - ： -  7586 
 [vagrant @ localhost〜] $ curl -o / dev / null https://google.com -w
％总计接收百分比％Xferd平均速度时间时间时间当前
载入上传总剩余速度
 110 220 110 220 0 0 42 0 0:00:05 0:00:05  - ： - ： -  7586 
 [vagrant @ localhost〜] $ curl -o / dev / null https://google.com -w
％总计接收百分比％Xferd平均速度时间时间时间当前
载入上传总剩余速度
 110 220 110 220 0 0 42 0 0:00:05 0:00:05  - ： - ： -  7857 
 [vagrant @ localhost〜] $ ping google.com 
 PING google .com（74.125.224.66）56（84）字节的数据。 
 64个字节从nuq04s07-in-f2.1e100.net（74.125.224.66）：icmp_seq = 1 ttl = 63时间= 7.64 ms 
 64个字节从nuq04s07-in-f2.1e100.net（74.125 .224.66）：icmp_seq = 2 ttl = 63 time = 6.26 ms 
来自nuq04s07-in-f2.1e100.net的64个字节（74.125.224.66）：icmp_seq = 3 ttl = 63 time = 8.69 ms 
 64 bytes from nuq04s07-in-f2.1e100.net（74.125.224.66）：icmp_seq = 4 ttl = 63 time = 7.89 ms 
 64 bytes from nuq04s07-in-f2.1e100.net（74.125.224.66）： icmp_seq = 5 ttl = 63 time = 5.99 ms 
来自nuq04s07-in-f2.1e100.net的64个字节（74.125.224.66）：icmp_seq = 6 ttl = 63 time = 8.64 ms 
  

对于记录，我有一个ubuntu 12.04.1虚拟机在同一个VirtualBox / OSX环境，不会遇到这个问题。 / p>

解决方案

确定，问题是DNS问题。

基本上CentOS VirtualBox已配置为使用本身内部的，但可能与我们网络上的某个内容冲突的10.0.2.3 DNS服务器。

在任何情况下，使用不同的DNS服务器解决了问题。

I'm running centos 6.3 inside of VirtualBox 4.2.1 on my OSX 10.8.2 machine, and have encountered a latency issue I do not comprehend. Basically, every http request to anywhere has an additional 5000ms delay. Ping has no additional delay.

[vagrant@localhost ~]$ curl -o /dev/null https://google.com -w ""
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
110   220  110   220    0     0     42      0  0:00:05  0:00:05 --:--:--  7586
[vagrant@localhost ~]$ curl -o /dev/null https://google.com -w ""
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
110   220  110   220    0     0     42      0  0:00:05  0:00:05 --:--:--  7586
[vagrant@localhost ~]$ curl -o /dev/null https://google.com -w ""
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
110   220  110   220    0     0     42      0  0:00:05  0:00:05 --:--:--  7857
[vagrant@localhost ~]$ ping google.com
PING google.com (74.125.224.66) 56(84) bytes of data.
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=1 ttl=63 time=7.64 ms
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=2 ttl=63 time=6.26 ms
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=3 ttl=63 time=8.69 ms
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=4 ttl=63 time=7.89 ms
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=5 ttl=63 time=5.99 ms
64 bytes from nuq04s07-in-f2.1e100.net (74.125.224.66): icmp_seq=6 ttl=63 time=8.64 ms

For the record, I have a ubuntu 12.04.1 virtual machine in the same VirtualBox/OSX environment, that does not experience the issue.

解决方案

Ok, it turns out the issue was a DNS issue.

Basically the CentOS VirtualBox was configured to use a 10.0.2.3 DNS server that was internal to itself, but maybe conflicted with something on our network.

In any case, using a different DNS server resolved the issue.

这篇关于CentOS VirtualBox对http请求的第二个延迟的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！