INTERACT FORUM
Linux => JRiver Media Center 32 for Linux => Topic started by: max096 on January 20, 2024, 07:18:32 pm
-
First MC32 images are now also available.
https://hub.docker.com/r/shiomax/jrivermc
-
Thank you.
I'll get on to that later today
-
Thanks Max, much appreciated!
-
I just tried to get MC32 by pulling shiomax/jrivermc:latest but ended up with 31.0.83 Did I do something wrong?
-
I just tried to get MC32 by pulling shiomax/jrivermc:latest but ended up with 31.0.83 Did I do something wrong?
Ill look into it later today should have been 32 already. You can use :latest-32 instead for now.
-
Thanks again Max.
I noticed some of the vnc settings have changed from my previous docker install, I've attempted to update them to work, however i seem to be stuck "connecting" when initiating a vnc session with the container. I have jriver running on a custom docker network to work with a reverse proxy so I found I needed ALLOW_ALL instead of LOCALHOST to at least get a password prompt, however I'm still stuck on 'connecting' after hitting connect (with no password entered).
Any ideas?
UID/GID
───────────────────────────────────────
USER_ID: 1000
GROUP_ID: 100
APP_USER: app
───────────────────────────────────────
setup-devices: no cdrom drive available: device /dev/sr0 not exposed to the container.
nginx: enabling https configuration...
prevent-minimize: starting...
nginx: waiting for certificate generation...
nginx: starting...
setup-app: Set config directory owner...
setup-app: VNC_LISTEN set to ALLOW_ALL. Allowing VNC connections from anywhere. This may be unsafe.
setup-intel-gpu: Intel GPU setup is blocked by INSTALL_INTEL_GPU_DRIVER env variable.
(https://i.ibb.co/StCwRLS/Screenshot-2024-01-28-at-12-47-32-pm.png)
-
Thanks again Max.
I noticed some of the vnc settings have changed from my previous docker install, I've attempted to update them to work, however i seem to be stuck "connecting" when initiating a vnc session with the container. I have jriver running on a custom docker network to work with a reverse proxy so I found I needed ALLOW_ALL instead of LOCALHOST to at least get a password prompt, however I'm still stuck on 'connecting' after hitting connect (with no password entered).
Any ideas?
Can you connect with the webui? VNC setup changed with the new images quite some time ago around mc29. If you set ALLOW_ALL you can connect from anywhere, but you currently still cannot set a password for direct VNC connection.
With the webui you will be asked to create a username/password combination on first visit and that will be used for login there. You also do not need ALLOW_ALL there. By default VNC is not reachable at all and you can only connect to the webui.
-
Oh wow, I hadn't even noticed the webui, thanks makes a lot more sense now, thanks.
-
Ill look into it later today should have been 32 already. You can use :latest-32 instead for now.
I tried it again with :latest and stll had 31 but latest-32.0.9-v.04.2 worked fine. I now have 32 running on my server and my client
-
I tried it again with :latest and stll had 31 but latest-32.0.9-v.04.2 worked fine. I now have 32 running on my server and my client
I don't want to confuse things but I pulled
shiomax/jrivermc:latest
with docker-compose and a my yaml file and MC32.0.9 installed properly on my Synology NAS a few days ago.
-
I don't want to confuse things but I pulled
shiomax/jrivermc:latest
with docker-compose and a my yaml file and MC32.0.9 installed properly on my Synology NAS a few days ago.
I expect it's me...I pulled the latest image but was trying to use a container that was cloned from my MC31 container.
I'll get back to it in a few days and get it sorted
-
I expect it's me...I pulled the latest image but was trying to use a container that was cloned from my MC31 container.
I'll get back to it in a few days and get it sorted
Ah, that explains it. I tried that, too at first and it also stayed with MC31.
I guess updating to a new major version requires "starting from scratch" so to speak.
Using docker-compose via CLI works pretty well for me. For updates within a major version I use Portainer and it takes only a minute to update.
Thanks to Max who has been very helpful to everyone.
-
I have now done a complete clean install to a new container and then restored a back up from 31.
This works fine and I can connect from my Windows client.
However, my NAS is set to shut down at night when no one is using it. The following day when the shiomax container restarts I am getting the following error in the log.
useradd: user 'abc' already exists
followed by
s6-rc:warning: unable to start service setup-app: command exited 9
The MC server doesn't start. I can connect to https:[server ip]/5800 but just get a black panel with "Not Connected" and a Reconnect button that does nothing.
Does anyone have any ideas what is going on here?
-
There also seems to be a lag between updates of different images. For example shiomax/latest-32.0.12-v0.4.2 (which is really the latest from JRiver) is available, whereas shiomax/latest is currently still pulling 32.0.9
-
I have now done a complete clean install to a new container and then restored a back up from 31.
This works fine and I can connect from my Windows client.
However, my NAS is set to shut down at night when no one is using it. The following day when the shiomax container restarts I am getting the following error in the log.
useradd: user 'abc' already exists
followed by
s6-rc:warning: unable to start service setup-app: command exited 9
The MC server doesn't start. I can connect to https:[server ip]/5800 but just get a black panel with "Not Connected" and a Reconnect button that does nothing.
Does anyone have any ideas what is going on here?
Have you tried to restart the Container from within the Docker Container Manager? That sometimes helps me when things go wrong.
Also, only somewhat related, a NAS is really supposed to run 24/7 and powering it down and up again every day puts unnecessary stress on the hardware, especially the drives. My personal recommendation would be to keep the NAS running at all times.
-
Have you tried to restart the Container from within the Docker Container Manager? That sometimes helps me when things go wrong.
Also, only somewhat related, a NAS is really supposed to run 24/7 and powering it down and up again every day puts unnecessary stress on the hardware, especially the drives. My personal recommendation would be to keep the NAS running at all times.
Stopping and starting the container made no difference.
I take your point about the NAS being on all the time, but I don't think that is a cause of this problem. Previous versions of MC shiomax docker images have been fine.
I'm puzzled by the log entry about the 'abc' user. I have no users called abc
-
Stopping and starting the container made no difference.
I take your point about the NAS being on all the time, but I don't think that is a cause of this problem. Previous versions of MC shiomax docker images have been fine.
I'm puzzled by the log entry about the 'abc' user. I have no users called abc
I hear you. I agree that restarting the container should not lead to this issue. I was as baffled as you about that error. No 'abc' user here either.
EDIT: Should have mentioned that when I tried to update to MC32 with Portainer, I got the same errors.
Upgrading the 'long way", using CLI docker-compose did the trick but I don't want to use this for every minor update. Portainer is just so easy. Have you tried Portainer?
-
I have now done a complete clean install to a new container and then restored a back up from 31.
This works fine and I can connect from my Windows client.
However, my NAS is set to shut down at night when no one is using it. The following day when the shiomax container restarts I am getting the following error in the log.
useradd: user 'abc' already exists
followed by
s6-rc:warning: unable to start service setup-app: command exited 9
The MC server doesn't start. I can connect to https:[server ip]/5800 but just get a black panel with "Not Connected" and a Reconnect button that does nothing.
Does anyone have any ideas what is going on here?
I'm experiencing the same.
-
Stopping and starting the container made no difference.
I take your point about the NAS being on all the time, but I don't think that is a cause of this problem. Previous versions of MC shiomax docker images have been fine.
I'm puzzled by the log entry about the 'abc' user. I have no users called abc
Not sure why you are getting an error about it exactly. But the abc user will be gone again next update. I wanted to simplify user creation in the container and took some of what linuxserver io is doing as a baseline. They donīt use the group for anything appearently and just put abc as the primary group (pretty sure every user needs a primary group). I did intend to "remove that later" but forgot about it.
It should not cause any problems though. The group just is there in the current image, but is not used anywhere. JRiver is started with the PID/GID configured and not with the primary group of the user that was created.
:latest is also mc32. If itīs not you just have to pull the image before redeploying the container. Iīd say the latest-32 etc. tags are better to use for most cases so you donīt upgrade past major versions of mc on accident. But otherwise those are the same for mc32 right now (until there is a 33).
-
I'm experiencing the same.
Can you try again now? Should be fixed, I hope. Not sure why this would only happen for some. I never got this error.
-
Can you try again now? Should be fixed, I hope. Not sure why this would only happen for some. I never got this error.
I've tried again, from scratch, with latest-32
It installs OK, I restore my library and licence key, I can connect from my client PC and play music but if I restart the NAS I get the following errors in the containter log
shiomax-jrivermc-latest32
date stream content
04/02/2024 14:29 stdout s6-rc: warning: unable to start service setup-app: command exited 127
04/02/2024 14:29 stdout /etc/s6-overlay/s6-rc.d/setup-app/run: 51: usemod: not found
-
I've tried again, from scratch, with latest-32
It installs OK, I restore my library and licence key, I can connect from my client PC and play music but if I restart the NAS I get the following errors in the containter log
shiomax-jrivermc-latest32
date stream content
04/02/2024 14:29 stdout s6-rc: warning: unable to start service setup-app: command exited 127
04/02/2024 14:29 stdout /etc/s6-overlay/s6-rc.d/setup-app/run: 51: usemod: not found
That explains a lot. Not the error (that's just another typo -.- should be fixed too now) but why I would have not seen it. I basically never restart my NAS / Server and donīt really stop docker containers either.
-
Thank you max! It's all good now!
I'm now trying to get the browser to work to enable Last.FM functionality, but nothing comes up. I've got the following environmental variable set in my docker compose yaml file:
- INSTALL_BROWSER=1
-
Thank you max! It's all good now!
I'm now trying to get the browser to work to enable Last.FM functionality, but nothing comes up. I've got the following environmental variable set in my docker compose yaml file:
- INSTALL_BROWSER=1
Like stated in the patch notes from v0.3.0 for some reason it does not work for last.fm right now. It opens some links in the browser but not all of them. Itīs still the same as it was. Would need to figure out where I need to add firefox to for mc to find it.
-
It's working for me now. I have started fresh with 32-latest, restored my library and then successfully stopped and restarted the container. Thank you Max
-
I have installed the M32 docker image on my Raspberry Pi4 and it is running normally, according to Portainer. However, when I try to VNC into it (either with VNC Viewer or MacOS Screen Sharing) using the <IP-address:5900> I only get the rPi desktop, not MC32. What am I doing wrong?
Many thanks for any help.
-
I have installed the M32 docker image on my Raspberry Pi4 and it is running normally, according to Portainer. However, when I try to VNC into it (either with VNC Viewer or MacOS Screen Sharing) using the <IP-address:5900> I only get the rPi desktop, not MC32. What am I doing wrong?
Many thanks for any help.
You probably already have another vnc running on your rpi. You can check if a port is in use by running 'sudo netstat -tulpn | grep LISTEN'. Probably first stop/remove the container. If thatīs the case you can use bridge networking and map vnc to 5801 instead.
Remove this from your config if you have it in there
network_mode: host
Add the ports
ports:
- "5801:5800"
- "5900:5900"
- "52100:52100"
- "52101:52101"
- "52199:52199"
- "1900:1900/udp"
Then you should have 5801 be VNC in the container and 5800 be your rpi still.
-
Thanks Max, I'll try that.
Interestingly, I found that if I VNC into the Pi via port 5900 and then open MC32 like I did when I connected without the port specification to open the native MC32 it is actually the docker MC32 that opens.
Clearly, this is only playing around because both the native version and the docker version on a Pi4, even at 2GHz, are painfully slow. They do work OK serving MC32 remotely so that's why I am playing with it. To have a backup MC32 server.
I just got a Pi5 that I still have to put together and want to see if that's significantly faster.
Many thanks again for your help! Much appreciated.
-
Thanks Max, I'll try that.
Interestingly, I found that if I VNC into the Pi via port 5900 and then open MC32 like I did when I connected without the port specification to open the native MC32 it is actually the docker MC32 that opens.
Clearly, this is only playing around because both the native version and the docker version on a Pi4, even at 2GHz, are painfully slow. They do work OK serving MC32 remotely so that's why I am playing with it. To have a backup MC32 server.
I just got a Pi5 that I still have to put together and want to see if that's significantly faster.
Many thanks again for your help! Much appreciated.
RPI5 uses a kernel by default that enabled 16k pages. The docker image does not work with that as of now.
You can put this into your boot.txt and then it should work.
kernel=kernel8.img
-
Thanks Max! Will do.
-
I installed MC32 on the Pi5 (running 64 bit Bookworm/Xfce) in bridge mode but I must have a permission problem because when I try to VNC into the container I get "The connection was refused by the computer" error.
I'll do some research to see if I can fix that.
-
I installed MC32 on the Pi5 (running 64 bit Bookworm/Xfce) in bridge mode but I must have a permission problem because when I try to VNC into the container I get "The connection was refused by the computer" error.
I'll do some research to see if I can fix that.
If youīre using bridge networking you need to set it to ALLOW_ALL. Because the container has its own localhost then, so restricting connections to localhost will make it so you can only connect from inside the container. Which isnīt all that straight forward to accomplish.
-
If youīre using bridge networking you need to set it to ALLOW_ALL. Because the container has its own localhost then, so restricting connections to localhost will make it so you can only connect from inside the container. Which isnīt all that straight forward to accomplish.
Thanks Max, I have ALLOW_ALL in my yaml but can't get into the container with VNC. No problem, though, I got native MC32 running rather quickly so I will probably not use the docker container on the pi5.
-
I have tried to update to MC 32.0.23 on my Synology DS1918+ using docker-compose pulling this image
shiomax/jrivermc:latest-32.0.23-v0.4.4
the container runs, according to the container manager and Portainer but when I try to access via VNC I get this error
The connection closed unexpectedly.
It's probably a Synology problem but I am stumped.
Would very much appreciate any ideas/hints on how to solve this.
-
I have tried to update to MC 32.0.23 on my Synology DS1918+ using docker-compose pulling this image
shiomax/jrivermc:latest-32.0.23-v0.4.4
the container runs, according to the container manager and Portainer but when I try to access via VNC I get this error
The connection closed unexpectedly.
It's probably a Synology problem but I am stumped.
Would very much appreciate any ideas/hints on how to solve this.
Not sure, just tried. This works with the latest image. Both web and vnc. It worked before? Try to restart the system if you did not yet. Might just fix it.
Otherwise, there are logs for tigervnc in /log/tigervnc. The file 'current' contains the latest logs.
The following will follow the logs live in your terminal
docker exec -it <container_name or id> /bin/bash
root@stuffs:/tmp# cd /log/tigervnc/
root@stuffs:/log/tigervnc# tail -f current
Then you can try to connect to it and see if any errors pop up. If nothing comes up at all youīve got some networking issue and donīt even make it to the container. Port 5900 has to be accessable and you have to have ALLOW_ALL set to be able to connect from other systems.
-
Not sure, just tried. This works with the latest image. Both web and vnc. It worked before? Try to restart the system if you did not yet. Might just fix it.
Otherwise, there are logs for tigervnc in /log/tigervnc. The file 'current' contains the latest logs.
The following will follow the logs live in your terminal
docker exec -it <container_name or id> /bin/bash
root@stuffs:/tmp# cd /log/tigervnc/
root@stuffs:/log/tigervnc# tail -f current
Then you can try to connect to it and see if any errors pop up. If nothing comes up at all youīve got some networking issue and donīt even make it to the container. Port 5900 has to be accessable and you have to have ALLOW_ALL set to be able to connect from other systems.
Thank you, Max
this is an excerpt of the log. Looks like the VNC server error repeats every second when the server is trying to restart
2024-03-10 14:46:52.970389689 Sun Mar 10 10:46:52 2024
2024-03-10 14:46:52.970391871 vncext: VNC extension running!
2024-03-10 14:46:52.970895944 (EE)
2024-03-10 14:46:52.970899500 Fatal server error:
2024-03-10 14:46:52.970902395 (EE) vncExtInit: unable to set socket to listening mode: Address already in use (98)
2024-03-10 14:46:52.970903788 (EE)
2024-03-10 14:46:53.978473000 Starting tigervnc...
2024-03-10 14:46:53.992089419
2024-03-10 14:46:53.992093450 Xvnc TigerVNC 1.12.0 - built 2023-01-06 16:01
2024-03-10 14:46:53.992096072 Copyright (C) 1999-2021 TigerVNC Team and many others (see README.rst)
2024-03-10 14:46:53.992098287 See https://www.tigervnc.org for information on TigerVNC.
2024-03-10 14:46:53.992100236 Underlying X server release 12101006, X.Org
2024-03-10 14:46:53.992101515
2024-03-10 14:46:53.992893797
2024-03-10 14:46:53.992896856 Sun Mar 10 10:46:53 2024
2024-03-10 14:46:53.992898833 vncext: VNC extension running!
2024-03-10 14:46:53.993568852 (EE)
2024-03-10 14:46:53.993571734 Fatal server error:
2024-03-10 14:46:53.993574311 (EE) vncExtInit: unable to set socket to listening mode: Address already in use (98)
2024-03-10 14:46:53.993575599 (EE)
2024-03-10 14:46:55.000237602 Starting tigervnc...
2024-03-10 14:46:55.013460387
2024-03-10 14:46:55.013466446 Xvnc TigerVNC 1.12.0 - built 2023-01-06 16:01
2024-03-10 14:46:55.013469025 Copyright (C) 1999-2021 TigerVNC Team and many others (see README.rst)
2024-03-10 14:46:55.013471276 See https://www.tigervnc.org for information on TigerVNC.
2024-03-10 14:46:55.013473232 Underlying X server release 12101006, X.Org
I haven't restarted the NAS yet because it's performing data scrubbing but will do so as soon as it's done.
EDIT: I have now restarted the NAS but the problem persists
EDIT2: SOLVED
When I tried to install the new version I had screwed up the web server on the NAS and that hugged all the ports. I un-installed the web server and all is well again