mpd creating lots of local connections?

coralfang · 04-22-2018, 02:12 AM

Any ideas why netstat gets spammed with TIME_WAIT connections for the mpd daemon?
I'm wondering if i have something configured wrong, and that this spam isn't supposed to be like this?

Code:

$ netstat -tupn
(Not all processes could be identified, non-owned process info
 will not be shown, you would have to be root to see it all.)
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 127.0.0.1:6600          127.0.0.1:56042         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:55992         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56006         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56020         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:55994         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56044         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:55998         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56038         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56048         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56014         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56052         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56000         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56012         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56024         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:48140         ESTABLISHED 29057/mpd           
tcp        0      0 127.0.0.1:6600          127.0.0.1:56040         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56050         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56010         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56028         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56046         TIME_WAIT   -                   
tcp        0      0 10.10.60.66:52532       64.30.224.237:80        ESTABLISHED 28697/mpdscribble   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56016         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56026         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:48140         127.0.0.1:6600          ESTABLISHED 28697/mpdscribble   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56022         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56036         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56002         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56008         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56054         TIME_WAIT   -                   
tcp        0      0 10.10.60.66:58698       64.30.224.237:80        ESTABLISHED 28697/mpdscribble   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56004         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56032         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:55996         TIME_WAIT   -                   
tcp        0      0 127.0.0.1:6600          127.0.0.1:56018         TIME_WAIT   -

Everything on port 6600 is the mpd daemon.

I run mpd once from my user login (checks if it is already running) and my mpd.conf is:

Code:

$ cat  ~/.config/mpd/mpd.conf 
music_directory		"~/Music"
playlist_directory	"~/.config/mpd/playlists"
db_file				"~/.config/mpd/database"
log_file			"~/.config/mpd/log"
pid_file			"~/.config/mpd/pid"
state_file			"~/.config/mpd/state"
sticker_file		"~/.config/mpd/sticker.sql"
port				"6600"
restore_paused "yes"
auto_update	"yes"

input {
        plugin "curl"
}

#pulseaudio
audio_output {
	type		"pulse"
	name		"pulse audio"
}

#visualizer
audio_output {
        type            "fifo"
        name            "my_fifo"
        path            "/tmp/mpd.fifo"
        format          "44100:16:2"
}

replaygain			"album"
#replaygain_preamp		"0"
#replaygain_missing_preamp	"0"
#replaygain_limit		"yes"
volume_normalization		"yes"
filesystem_charset		"UTF-8"

MensaWater · 04-23-2018, 03:09 PM

I don't know the answer but noticed most of what you're seeing are from IP 127.0.0.1 to 127.0.0.1 which is a/k/a "localhost". This means these are internal connections so aren't accessible outside the system (as opposed to the ones you see ESTABLISHED on your 10.x.x.x address). I wouldn't worry about them unless they continue to grow.

I found a discussion online where someone was suggesting using AF_UNIX sockets instead of these but that hadn't been implemented as recently as a month ago.

coralfang · 04-23-2018, 11:51 PM

Quote:

Originally Posted by MensaWater

I don't know the answer but noticed most of what you're seeing are from IP 127.0.0.1 to 127.0.0.1 which is a/k/a "localhost". This means these are internal connections so aren't accessible outside the system (as opposed to the ones you see ESTABLISHED on your 10.x.x.x address). I wouldn't worry about them unless they continue to grow.

I found a discussion online where someone was suggesting using AF_UNIX sockets instead of these but that hadn't been implemented as recently as a month ago.

Yeah, i'm aware it's all localhost. But as you just mentioned them growing, that's what seems to be happening. Say if i boot up the computer and check, there is only one or two connections with TIME_WAIT after mpd has started. Later, once the computer has been on all day, there could be 40-50 of them. Just curious why this happens.