TCPServer 错误: 已在使用的地址-bind (2)

几周前 Jekyll 对我来说还不错,但是现在它突然给了我以下错误:

TCPServer Error: Address already in use - bind(2)
INFO  WEBrick::HTTPServer#start: pid=7300 port=4000


% lsof -i :4000
<fetches nothing>

尽管端口上没有运行任何东西,下面是详细信息:

 % jekyll --version
Jekyll 0.11.2
% where jekyll
/home/bhaarat/.rvm/gems/ruby-1.9.2-p290/bin/jekyll
/usr/bin/jekyll
% ruby --version
ruby 1.9.2p290 (2011-07-09 revision 32553) [i686-linux]
% rvm --version
rvm 1.10.0

这是输出

 % jekyll --server
Configuration from /home/bhaarat/blog/omnipresent.github.com/_config.yml
Auto-regenerating enabled: /home/bhaarat/blog/omnipresent.github.com -> /home/bhaarat/blog/omnipresent.github.com/_site
[2012-04-21 13:46:40] regeneration: 38 files changed
[2012-04-21 13:46:40] INFO  WEBrick 1.3.1
[2012-04-21 13:46:40] INFO  ruby 1.9.2 (2011-07-09) [i686-linux]
[2012-04-21 13:46:40] WARN  TCPServer Error: Address already in use - bind(2)
[2012-04-21 13:46:40] INFO  WEBrick::HTTPServer#start: pid=7382 port=4000

我知道这个地址没有被使用 Jekyll 可能因为其他原因中断了,但是抛出了这个错误。我有什么选择?我也试过重新安装。

56614 次浏览

Type this in your terminal to find out the PID of the process that's using the 3000 port:

$ lsof -wni tcp:3000

Then, use the number in the PID column to kill the process:

$ kill -9 PID

Ctrl-Z doesn't terminate a program, but rather suspends it and sends it to the background. You can resume the program with the "fg" command. To actually terminate it, use Ctrl-C.

The actual error message seems to be bogus and can be ignored. I am getting the same error message "address in use" but jekyll works fine anyway at the expected port.

I have met this problem recently.

I tried out all the method mentioned above, and even restarted my computer, but still couldn't solve it!!! Then I removed the jekyll and installed a new version, it just worked.

gem uninstall jekyll & gem install jekyll (maybe you need super user priviledge).

If you really get annoyed with similar bugs, this sb method is worth a try...

work around

in /_site run: python -m SimpleHTTPServer 8080

Check that you don't have another terminal open where you are already running a server. If that is the case, do a CTRL-C to shutdown the server, and that will free the port/address.

I was not qualified to post comment. So I added a new answer.

I encountered this problem on Mac OS X 10.10.3. And I had never installed/used Jekyll before. I was not able to start jekyll server with its default port number 4000. The reason was that the port was the same as what NoMachine used. With

$ sudo lsof -wni tcp:4000

Note: Running this command without sudo will have no output.

I saw this output:

COMMAND PID USER   FD   TYPE             DEVICE SIZE/OFF NODE NAME
nxd     449   nx    3u  IPv4 0x8d22************      0t0  TCP *:terabase (LISTEN)
nxd     449   nx    4u  IPv6 0x8d22************      0t0  TCP *:terabase (LISTEN)

The port 4000 was occupied by nxd, which was the process started by NoMachine. And

$ sudo kill -9 449

would not work, because NoMachine's nxd process would keep restarting, with a new PID.

Therefore, I had to either:

  • Changed my jekyll server port in the site _config.yml to another spared one. I appended the line below to _config.yml and it worked.

    port: 3000 # change server port to 3000

or

  • Changed NoMachine's default nxd port, or Uninstall NoMachine

First you need to find PID of the process that's using the 3000 port:

 $ps -ef

Output Like this:

1003      4953  2614  0 08:51 pts/0    00:00:00 -bash
1003      5634     1  0 08:56 pts/0    00:00:00 spring server | moviestore | started 2 hours ago
1003      5637  5634  0 08:56 ?        00:00:01 spring app    | moviestore | started 2 hours ago | development mode
1003      6078  4953  0 09:03 pts/0    00:00:03 puma 3.6.0 (tcp://localhost:3000) [moviestore]
1003      6117  2614  0 09:03 pts/1    00:00:00 -bash
root      6520     2  0 09:57 ?        00:00:00 [kworker/u8:2]
root      6936  1225  0 11:09 ?        00:00:00 [lightdm] <defunct>
1003      7084     1  0 11:09 ?        00:00:00 /usr/bin/python /usr/share/apport/apport-gtk
1003      7475     1  0 11:10 ?        00:00:00 /usr/bin/python /usr/share/apport/apport-gtk
root      8739  1225  1 11:29 tty8     00:00:11 /usr/bin/X :1 -auth /var/run/lightdm/root/:1 -nolisten tcp vt8 -novtswitch
root      8853  1225  0 11:29 ?        00:00:00 lightdm --session-child 13 22
1002      8943     1  0 11:30 ?        00:00:00 /usr/bin/gnome-keyring-daemon --daemonize --login
1002      8954  8853  0 11:30 ?        00:00:00 gnome-session --session=ubuntu
1002      8992  8954  0 11:30 ?        00:00:00 /usr/bin/ssh-agent /usr/bin/dbus-launch --exit-with-session gnome-session --session=ubuntu
1002      8995     1  0 11:30 ?        00:00:00 /usr/bin/dbus-launch --exit-with-session gnome-session --session=ubuntu
1002      8996     1  0 11:30 ?        00:00:00 //bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session
1002      9007  8954  0 11:30 ?        00:00:00 /usr/lib/gnome-settings-daemon/gnome-settings-daemon
1002      9015     1  0 11:30 ?        00:00:00 /usr/lib/gvfs/gvfsd
1002      9018  8954  1 11:30 ?        00:00:07 compiz
1002      9021     1  0 11:30 ?        00:00:00 /usr/lib/x86_64-linux-gnu/gconf/gconfd-2
1002      9028  8954  0 11:30 ?        00:00:00 /usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1
1002      9029  8954  0 11:30 ?        00:00:01 nautilus -n
1002      9030  8954  0 11:30 ?        00:00:00 /usr/lib/gnome-settings-daemon/gnome-fallback-mount-helper
1002      9031  8954  0 11:30 ?        00:00:00 nm-applet
1002      9032  8954  0 11:30 ?        00:00:02 /opt/mTrac/mTrac
1002      9033  8954  0 11:30 ?        00:00:00 bluetooth-applet
1002      9045  9032  0 11:30 ?        00:00:00 /opt/mTrac/mTrac --type=zygote --no-sandbox
1002      9050     1  0 11:30 ?        00:00:00 /usr/lib/gvfs/gvfs-gdu-volume-monitor
1002      9054     1  0 11:30 ?        00:00:00 /usr/bin/pulseaudio --start --log-target=syslog
1002      9057     1  0 11:30 ?        00:00:00 /usr/lib/gvfs/gvfs-gphoto2-volume-monitor
1002      9062     1  0 11:30 ?        00:00:00 /usr/lib/gvfs/gvfs-afc-volume-monitor

here you can see :

1003      6078  4953  0 09:03 pts/0    00:00:03 puma 3.6.0 (tcp://localhost:3000) [moviestore]

localhost:3000 have pid: 6078

kill that process by

$sudo kill 6078

then run

$rails s

we can user fuser command

fuser -k 3000/tcp

Based on the top answer, I came up with a simple way

lsof -wni tcp:3000 | xargs -I{} kill -9 {}

It takes the output of finding PID in which port the tcp server is running at. In this case it is 3000.

It then kills the processes running that PID