Very easy. If you are comfortable with ssh, you add a .pls file with the connection information into the directory on the Pi SD card where the pre-defined stations have their pls files, or you can simply put the pls file in the directory where your own music files (NAS, SD card directory, or attached drive) are located. If you do the latter, the station you add will not be under the 'Radio' browse tab, it will be found in your music file directory structure.
Skip
at least I'm not alone. Also a friend with identical config as mine (RPi1B+) had this trouble before it happened to me.
All these cards corrupted? mmm no way.
This is way beyond my knowledge, can't help on that!
Sure the fact that has been reported only on ARMv6l (raspy 1) seems weird.
someone just reported a Segmentation fault issue on mpd bug tracker (also armv6l)
https://bugs.musicpd.org/view.php?id=4571
He attacked a gdb backtrace too.
Hi,
I think the segfault is due to bug in MPD 0.19.15. Next Moode update includes 0.19.17 so maybe no more segfaults. Btw, the most recent Moode update includes an enhanced watchdog monitor that will automatically restart MPD if it crashes.
Have u installed the most recent update?
-Tim
Hi,
U need to map a drive directly to USB/Disk_Label to avoid having incorrect free space reported. I think its a bug in Samba thats causing this issue.
For example if your USB disk is named MyUSBDisk then
z: --> \\moode\USB\MyUDBDisk
-Tim
Great!
Yes, if the most recent update is still 2016-07-28, then yes the system was up to date.
Also pretty sure mpd didn't restart by itself, and I spent like 5 minutes after the crash between sshing, reading/downloading the logs etc
at the end I had to restart the mpd.service manually.
So maybe watchdog didn't work there for some reason.
When the segfault happened these were the specs:
Hi,
Thats odd. Watchdog wakes up every 6 seconds and checks to see of MPD is in the process list. If not it writes an entry to Moode log and then restarts MPD. Look in /var/log/moode.log for "watchdog"
-Tim
I have this, currently...
Code:
pi@moodeBrev2:/var/log $ ps -A | grep w
5 ? 00:00:00 kworker/0:0H
11 ? 00:00:00 writeback
17 ? 00:00:00 kswapd0
57 ? 00:00:00 dwc_otg
62 ? 00:00:00 deferwq
85 ? 00:00:00 kworker/0:1H
618 ? 00:05:20 watchdog.sh
676 ? 00:31:48 worker.php
957 ? 00:00:00 kworker/u2:1
19962 ? 00:00:00 kworker/0:2
21293 ? 00:00:01 kworker/0:0
24865 ? 00:00:00 kworker/u2:0
26025 ? 00:00:00 kworker/u2:2
pi@moodeBrev2:/var/log $
Code:
pi@moodeBrev2:/var/log $ uname -a
Linux moodeBrev2 4.4.8+ #880 Fri Apr 22 21:27:42 BST 2016 armv6l GNU/Linux
pi@moodeBrev2:/var/log $...And this in moode.log...
Code:
20160820 091906 worker: wlan0 address not assigned
20160820 091906 worker: eth0 exists
20160820 091907 worker: eth0 (192.168.1.3)
20160820 091908 worker: NAS sources (mountall initiated)
20160820 091908 worker: MPD consume reset to off
20160820 091908 worker: End startup
20160820 091908 worker: Ready
20160824 221232 openMpdSocket(): could not connect to MPD
20160824 221232 openMpdSocket(): could not connect to MPD
20160824 221235 openMpdSocket(): could not connect to MPDSo Watchdog.sh is running but it doesn't seem to react to this....
Code:
Aug 24 22:11:54 moodeBrev2 dhcpcd[571]: eth0: Router Advertisement from fe80::8e10:d4ff:fe62:2a8e
Aug 24 22:12:32 moodeBrev2 systemd[1]: mpd.service: main process exited, code=killed, status=11/SEGV
Aug 24 22:12:32 moodeBrev2 systemd[1]: Unit mpd.service entered failed state.
Aug 24 22:12:38 moodeBrev2 dhcpcd[571]: eth0: Router Advertisement from fe80::8e10:d4ff:fe62:2a8eJ.
OK, from what little my brain can hold of shell programming.... It doesn't seem to restart mpd from watchdog.sh. Does this mean I am backlevel?
J.
Code:
pi@moodeBrev2:/var/www/command $ sudo find / -name watchdog.sh
/var/www/command/watchdog.sh
pi@moodeBrev2:/var/www/command $
Code:
#!/bin/bash
#
# This Program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3, or (at your option)
# any later version.
#
# This Program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Moode; see the file COPYING. If not, see
# <http://www.gnu.org/licenses/>.
#
# Moode Audio Player (C) 2014 Tim Curtis
# [url=http://moodeaudio.org]moodeaudio.org[/url]
#
# 2016-02-27 2.5 TC watchdog for too many child instances of engine
# 2016-06-07 2.6 TC moodeOS 1.0
#
LIMIT=18
COUNT=$(pgrep -c -f "php-fpm: pool display")
while true; do
if (( COUNT > LIMIT )); then
TIMESTAMP=$(date +'%Y%m%d %H%M%S')
LOGMSG=" watchdog: PHP restarted, fpm child limit "$LIMIT" exceeded"
echo $TIMESTAMP$LOGMSG >> /var/log/moode.log
systemctl restart php5-fpm
fi
sleep 6
COUNT=$(pgrep -c -f "php-fpm: pool display")
done > /dev/null 2>&1 &J.
OK, I've found how to grab the updater and updated my MoOde.
Wow that information wasn't easy to find. Even though I knew it was in this topic somewhere. Maybe there should be a MoOde audio player indexing topic? Or an update to the first post?
Just reviewed watchdog.sh and the command to monitor mpd is now in there.
Wow that information wasn't easy to find. Even though I knew it was in this topic somewhere. Maybe there should be a MoOde audio player indexing topic? Or an update to the first post?
Just reviewed watchdog.sh and the command to monitor mpd is now in there.
Hi,
Easiest to just follow Moode Twitter feed. Just scroll down a bit and you will see a post referencing Updater.
Moode 2.7 has Updater built in so once its released life will get a bit easier
-Tim
I think I've found why the watchdog is not working!
I made a test. I killed mpd manually.
Watchdog does :
Code:
pgrep -c -f mpdIf you do list instead of count you see:
Code:
pgrep -f -l mpd
1213 sudo
1224 mpdasBut since there is mpdas (mpd scrobbler) running there, the watchdog is happy with that, as it contains the string mpd, so it does nothing
If the watchdog did something like:
Code:
pgrep -c -f -x "/usr/local/bin/mpd --no-daemon /etc/mpd.conf"EDIT: I've just tried modding that string in watchdog.sh with mine and it works perfectly. I kill mpd and it restarts perfectly and adds the entry to moode.log.
Last edited:
Hi,
I've tested it on Pi-1B that is already running SoX resampling at 24/176.4 and no issues. My 1B's are overclocked to 1Ghz.
-Tim
I think I've found why the watchdog is not working!
I made a test. I killed mpd manually.
Watchdog does :
and it shows the count. It shows "2", but just by that you don't now the services names.Code:pgrep -c -f mpd
If you do list instead of count you see:
Remember I killed mpd. At that point you realize that it's listing "mpdas" entries, not "mpd"...Code:pgrep -f -l mpd 1213 sudo 1224 mpdas
But since there is mpdas (mpd scrobbler) running there, the watchdog is happy with that, as it contains the string mpd, so it does nothing
If the watchdog did something like:
then it would work, just tracking mpd and ignoring mpdasCode:pgrep -c -f -x "/usr/local/bin/mpd --no-daemon /etc/mpd.conf"
EDIT: I've just tried modding that string in watchdog.sh with mine and it works perfectly. I kill mpd and it restarts perfectly and adds the entry to moode.log.
Hi,
Nice catch!
Try pgrep -c -f "mpd "
I think this will also work.
-Tim
I didn't add this earlier as components are different etc....
I run a B rev 2 overclocked to 900MHz. Running Charlie Laub's ACD ALSA plugin set to provide baffle step compensation and a HF lift on my Aura NS3s in Woden Redeye speakers. This is running via a wired ethernet and works fine. If I had to be picky, the interface is sometimes a little slower to update but nothing glacially slow.
Noticed that there have been a further two releases to MPD for better stability in including a memory leak fix.
Now 0.19.19.
Sent from my P01M using Tapatalk
Hi,
Compiling...
.19 will be included in next Moode update
-Tim