Let me know if I should be asking for help under another discussion topic.

I've been unable to start LMS for many months, today I re-approached the issue by upgrading from rc.5 to 5.0.3.

During the upgrade I removed the /config/plugins folder from my flash drive. Once I verified that the server would boot up and run without a problem, I downloaded lms_overbyrn.plg and moved it to a new folder on my flash; /config/plugins.

After restarting the server, I found the LMS info screen was available, through the WEb gui, under settings (see screenshot). Enabling the LMS Service seemed to run successfully but, starting the LMS Daemon fails.

I suspect the root cause will be a previous LMS install from over a year ago, I may need to locate and remove the remains of that install.

Any ideas would be appreciated, I have 2 squeezebox radios I would like to use with Unraid.

having the same issues on a brand new v5 install

 

plugin installs but service won't start

 

tried every tip I can find and so far no joy

 

markiii / icelake, I've just tried the lms_overbyrn.plg plugin here on a fresh v5.0 system and it's installed and started fine.  My first suggestion would be to perform the installation and startup all from command line where you'll get to see if anything funky is happening.

 

Assuming you can afford to start off with a system that doesn't have LMS installed at all, then try doing something like this;

 

1. remove the lms_overbyrn.plg from the plugins dir and reboot

2. put the lms_overbyrn.plg file back into plugins dir

3. fire up a command line and install the plugin using "installplg /boot/config/plugins/lms_overbyrn.plg"

4. watch what it does and note any issues along the way.  It should however install the plugin without issue.  At this point LMS itself is not yet downloaded or installed.

5. Go to LMS plugin settings page and configure your location, port etc.

6. Do NOT press "Install" button and do NOT set the service to enable.  Just change the location, port and branch and hit apply.

7. Back at command line, issue the following command; "/etc/rc.d/rc.lms buttonstart"

8. Again note down what it's doing and pay attention to any red flags.  The output should look like this;

root@Tower:~# /etc/rc.d/rc.lms buttonstart

Installing LMS...

No local downloaded archive in plugin dir, checking online for latest version...

Latest stable version : 7.7.3

Downloading logitechmediaserver-7.7.3.tgz to /boot/config/plugins/lms...

Download complete...

Extracting LMS application to /tmp...

Cleaning unnecessary files...

Moving LMS application to /mnt/disk1/lms...

Creating default playlist directory (/mnt/disk1/lms/playlists)

Changing ownership (nobody) & permissions (0777 rwxrwxrwx)

... OK

Starting LMS...

... OK

 

9. Check that the LMS process is running by typing "ps -ef | grep -v grep | grep slimserver"

You should get back one line like ;

nobody    4985    1  0 20:51 pts/0    00:00:01 /usr/bin/perl /mnt/disk1/lms/slimserver.pl --dameon --user nobody --group users --httpport 9000 --pidfile /var/run/lms/lms.pid

--cachedir /mnt/disk1/lms/Cache --prefsdir /mnt/disk1/lms/prefs > /dev/null 2>&1

 

Assuming you get this far, then try to open up the LMS web page.  eg. http://tower:9000

 

It should work...

 

 

Thanks will give that a try,

 

What's the difference between that and doing it from the webgui, that makes you think it will fix it?

 

on a hunch it was a permissions issue, and before trying your solution, I decided to completely start afresh, new precleared drives, new v5 install, complete from scratch

 

result, everything installed and started

 

so next I tried enabling nightly builds

 

and straight away got the issue with the lms.pid again

 

reverted to stable, and its fine again, so it looks like something with the latest 7.8 nightly build does not want to install. Not a great issue as the 7.3 stable is an improvement on the 7.2 running on my production box anyway

 

but odd none the less

I followed the directions overbyrn gave in his post above.  When I grepped to see if slimserver was running I see that it is. However, when I go to tower:9000 I don't see the LMS control page.  Does a webserver need to be configured for LMS to be able to put up the page?

on a hunch it was a permissions issue, and before trying your solution, I decided to completely start afresh, new precleared drives, new v5 install, complete from scratch

 

result, everything installed and started

 

so next I tried enabling nightly builds

 

and straight away got the issue with the lms.pid again

 

reverted to stable, and its fine again, so it looks like something with the latest 7.8 nightly build does not want to install. Not a great issue as the 7.3 stable is an improvement on the 7.2 running on my production box anyway

 

but odd none the less

markiii, thanks for the update.  I didn't try to use the nightly build in my test.  Perhaps something has changed as you say.  Presumably by the time you came to try to nightly, you'd already had the stable running?  I ask so that I can fairly accurately replicate your install steps this end.

 

I followed the directions overbyrn gave in his post above.  When I grepped to see if slimserver was running I see that it is. However, when I go to tower:9000 I don't see the LMS control page.  Does a webserver need to be configured for LMS to be able to put up the page?

Nope, doesn't need a separate web server. Everything is self contained within LMS.

 

If the process is running, I'd expect it to be available on the hostame:port you have configured.  Can you reach it by IP:PORT instead? 

 

When you grep for the process and it returns the details, check the parameter "--httpport" is using the port you think it is.

 

eg.

nobody    4985    1  0 Dec04 ?        00:00:27 /usr/bin/perl /mnt/disk1/lms/slimserver.pl --dameon --user nobody --group users --httpport 9000 --pidf

ile /var/run/lms/lms.pid --cachedir /mnt/disk1/lms/Cache --prefsdir /mnt/disk1/lms/prefs > /dev/null 2>&1

 

 

Maybe try to see if the port is listening;

netstat -anp| grep LISTEN

 

You should get a list back with probably three entries where perl is running;

root@Tower:~# netstat -anp | grep LISTEN

tcp        0      0 0.0.0.0:9090            0.0.0.0:*              LISTEN      4985/perl

tcp        0      0 0.0.0.0:37              0.0.0.0:*              LISTEN      1116/inetd

tcp        0      0 0.0.0.0:9000            0.0.0.0:*              LISTEN      4985/perl

tcp        0      0 0.0.0.0:50155          0.0.0.0:*              LISTEN      4985/perl

tcp        0      0 0.0.0.0:139            0.0.0.0:*              LISTEN      1617/smbd

tcp        0      0 0.0.0.0:111            0.0.0.0:*              LISTEN      1098/rpc.portmap

tcp        0      0 0.0.0.0:80              0.0.0.0:*              LISTEN      1584/emhttp

tcp        0      0 0.0.0.0:21              0.0.0.0:*              LISTEN      1116/inetd

tcp        0      0 0.0.0.0:50582          0.0.0.0:*              LISTEN      1105/rpc.statd

tcp        0      0 0.0.0.0:23              0.0.0.0:*              LISTEN      1116/inetd

tcp        0      0 0.0.0.0:3483            0.0.0.0:*              LISTEN      4985/perl

tcp        0      0 0.0.0.0:445            0.0.0.0:*              LISTEN      1617/smbd

 

and if it shows in the list that is returned, try to telnet to it.  If telnet at least connects, then you know for sure it's listening and the problem is elsewhere.

Thanks Overbyrn.  When I checked for which ports were listening I found that VBoxHeadless was listening on port 9000. I changed virtualbox's port to 9001 and now lms starts up fine.

 

Anyone who is running the Virtualbox plugin and this LMS plugin should be aware that virtualbox plugin sets port 9000 as the default port for remote desktop connections and that this will conflict with the default port for LMS.

on a hunch it was a permissions issue, and before trying your solution, I decided to completely start afresh, new precleared drives, new v5 install, complete from scratch

 

result, everything installed and started

 

so next I tried enabling nightly builds

 

and straight away got the issue with the lms.pid again

 

reverted to stable, and its fine again, so it looks like something with the latest 7.8 nightly build does not want to install. Not a great issue as the 7.3 stable is an improvement on the 7.2 running on my production box anyway

 

but odd none the less

markiii, thanks for the update.  I didn't try to use the nightly build in my test.  Perhaps something has changed as you say.  Presumably by the time you came to try to nightly, you'd already had the stable running?  I ask so that I can fairly accurately replicate your install steps this end.

 

tried it both ways,

 

if I get stable running and then switch to nightlys, it tanks

 

If I start straight away with nightlys, it tanks

 

thanks for investigating

 

Thanks for the response, I followed the directions for moving the overbryn plugin and reinstalling lms from the command line. I saw no red flags, until the end when output said "starting LMS" instead of starting, output said "lms pid not created for some reason"

 

Not sure if this is related but, under shares, I have two folders, one "lms" and one "slimserver"

 

Thanks for any suggestions

I worked through the steps overbryn provided above. Like icelake I received no red flags until "lms.pid not created for some reason". Does anyone have any idea what might be going on?

 

I have attached the output from the console in case it helps

lms_-_pid_not_created_for_some_reason.txt

Same issue here : I had to rebuild-tree one of my disk (where my musics are but not the LMS installation) and since then, lms won't start anymore.

 

I tried to switch to nightlies then back to stable but no love came from that

 

I get the same "lms.pid not created for some reason" , no lms pid is running, 9000 port is free...

I don't get it.

 

Please help.

I got it working in the end, but I'm not sure what did it.

 

I tried a clean USB stick with a fresh install of unraid and no other plugins which resulted in the same "lms.pid not created for some reason" message, I think I also tried a few other things which appeared to make no difference and resulted in the same message.

 

Eventually I tried installing it to the default location on disk 1 instead of to the cache drive, to my surprise this worked. I then removed the installation and tried installing to the cache drive again, this time it work fine. I have powered down/up several times since and LMS starts up okay each time.

Eventually I tried installing it to the default location on disk 1 instead of to the cache drive, to my surprise this worked. I then removed the installation and tried installing to the cache drive again, this time it work fine. I have powered down/up several times since and LMS starts up okay each time.

 

This happened me several months ago as well. Constant pid error when installed to cache but works perfectly installed to Disk1. (I didn't try reinstalling to cache afterwards - leave well enough alone was my philosophy).

 

Peter

Well I reinstalled from scratch the LMS server on unraid (deleting the install folder on my cache drive under .apps and the lms folder in the plugin directory) and restored some files from the backups I made (/slimserver/cache folder, the one containing the databases of the previous install).

 

All is good now and I didn't have to reconfigure the whole server part nor rescan my whole music folders.

I've made a note of the pid not created issue that some have reported.  Unfortunately, plugin development must take a back-seat as I'm in the process of emigrating from UK to US and have way too many other things contending for my time over the next 4-6 months.

 

Darts:

 

Bit of a newbie, but have encountered the same problem... would you mind terribly setting out a step by step of your fix for this?  Can't find the .apps directory you mention above...

 

Any help greatly appreciated - house is v. quiet without lms!

My my I'm terribly sorry, it's been quite some time since I've been here.

Do you still need a hand?

 

The .apps is a floder I created myself on the cache disk in order to regroup all my apps appart from the cached files (that will be moved by the mover script every once in a while) .

Adding a dot before the name of the folder makes it "invisible" and prevents the mover to see it.

 

So basically when I say "deleting the install folder on my cache drive under .apps" I'm merely saying "delete the folder created by the plugin's installation" (the one you can put anywhere in your lms settings) not necessarily the one on your flash drive.

 

I'll keep an eye on this post in the next few days.

I'm running into this issue as well.  It had been working on unRaidv5, but once I upgraded to unRaidv6b5 I'm getting the pid error.  I turned off all the plugins, tried manually installing, tried to install to disk1 directly but I still get the error.  I've tried 7.8 stable build, 7.8 nightly, 7.9...each time deleting the install folder and starting fresh.

 

Any other ideas?

Damn! I'm getting the error as well, no way to get back a working one

Has anyone figured out how to use a stable 7.8 version on unraid?

Is there another plugin than Overbyrn's who unfortunately doesn't seem to have the time to take care of it anymore ?

I wish someone could help with this. I just cant get past the pid error.