[Support] Joshndroid's Docker Repo Support Thread


Recommended Posts

8 minutes ago, Nesquik said:

Thanks Josh! I will keep it on the LSIO network since I am going to put it behind the reverse proxy. 

 

Do most of the features work? I don't think I could get it to login. My goal is to get my YouTube subscriptions loaded.

 

I was able to create a user and start subscribing to channels. I can log into this same user across devices which works just fine. I havn't had much need to do much else apart from that and then muck around with the resolution stuff and try out a couple downloads which seems to all work.

 

On mobile I just create a link on my homescreen that launches the browser as a 'browser based app' sort of thing. It is a tad clunky on mobile but meh, I've basically deleted youtube entirely now.

 

I do get a bit of buffering at times which seems to buffer for eternity. That may be a settings issue, that may be an internet issue, I am yet to work that out. I typically have to refresh the page and then play the video again, jumping to where I was. I need more time to look at the settings and what I can do to improve that. there is also the possibility that my internet cant handle it (its garbage)

  • Like 1
Link to comment

Can we not just include the database inside the docker itself so end users and click and install?  This is a crazy amount of editing and if their was an update I suspect everything to break and an endless amount of requests to fix it.  It seems much easier to just have it already setup in the docker.  (invidious)

Edited by kilobit
Link to comment
On 8/30/2021 at 2:52 AM, kilobit said:

Can we not just include the database inside the docker itself so end users and click and install?  This is a crazy amount of editing and if their was an update I suspect everything to break and an endless amount of requests to fix it.  It seems much easier to just have it already setup in the docker.  (invidious)

 

Been using this for months at this point with updates almost daily and nil issue.

 

All the SQL commands are within the files as noted. If something changes at this point I have more people able to not only understand the commands but also where they are. I've already had good people help with the template/setup in getting this far. I suspect that you may be able to script the commands to be one line which also may make it quicker/easier.

 

Nothing stopping you or anyone else developing an all in one docker on ubuntu base or something. This one could be the minimalist docker for those who want to run their own database.

Link to comment

I got invidious running great locally 192.x.x.x but I am unable to get it working on NGINX proxy manager.  Anyone got it working and mind sharing your config?  I have other stuff like jellyfin and Bitwarden working so I have a fairly good understanding on NGINX proxy manager.

 

Ideally I'd like to redirect it to a full base domain  like.   https://example.com.    previously all of my reverse proxied services are subdomains like Bitwarden.example.com

 

Link to comment
  • 2 weeks later...
On 9/3/2021 at 3:54 AM, thunderclap said:

I'm just becoming aware of libreddit. Is there a way to log into my account from it so the front page shows what I'm interested in?

 

Nah you are unable to do that at this time.

You also obviously can't reply.

So i kind of run this in tandem with an ad free reddit client.

Basically you have to go through you subscriptions and 'subscribe' within libreddit to be able to generate a bit of a feed.

 

Once I did that, I went to settings and at the bottom there is a link called 'You can restore your current settings and subscriptions after clearing your cookies using this link.' the 'think link' is a good little hyperlink that i have copied and saved as my launching link in heimdall (or whatever dashboard your using. it means that I get the fresh experience each time.

 

On 9/7/2021 at 4:37 AM, darkrom said:

I got invidious running great locally 192.x.x.x but I am unable to get it working on NGINX proxy manager.  Anyone got it working and mind sharing your config?  I have other stuff like jellyfin and Bitwarden working so I have a fairly good understanding on NGINX proxy manager.

 

Ideally I'd like to redirect it to a full base domain  like.   https://example.com.    previously all of my reverse proxied services are subdomains like Bitwarden.example.com

 

 

I have this running on nginx proxy manager (both the linuxserver verison and within the official docker) however I don't have this running as a full base domain, only as a invidious.example.com which is working exactly as intended the multiple times I have set this up. Typically i dont have my things exposed to the outside world so that entails that I have the invidious.example.com setup as a local DNS entry in pihole such that it points to the local device running nginx proxy manager 192.168.0.xx.... then it will stay local, hit the proxy manager and reverse proxy to the correct container which is under a custom network 'proxnet' so 172.20.0.xx or whatever with the port as necessary (default is 3000)

 

Link to comment

invidious instructions

Update - Made a small edit to the SQL code block and instructions regarding using the entire code block in one go.

Tested on a fresh database setup today I was able to copy the entire code block in one and paste it into the terminal, without the necessity of doing paragraph by paragraph (for anyone wondering).

Edited by Joshndroid
Link to comment
  • 3 weeks later...

regarding invidious setup problems:
After I got past the postgres hostname problem by adding a dns entry to my pi-hole and using --dns as extra parameter for this container the logs stated that the password for my DB was incorrect. As I could login manually I had a suspicion and changed the db users password to "kemal" and that did the trick so I investigated and found that the config path of the unraid template is wrong. It is mapped to /config in the container but invidious uses /invidious/config. That's the reason it's looking for postgres as host even when you specified an IP in your config file and tries to connect to the db with the default "kemal" as password simply because it uses the example config instead of your customized one.

Edited by Netti93
Link to comment
3 hours ago, Netti93 said:

regarding invidious setup problems:
After I got past the postgres hostname problem by adding a dns entry to my pi-hole and using --dns as extra parameter for this container the logs stated that the password for my DB was incorrect. As I could login manually I had a suspicion and changed the db users password to "kemal" and that did the trick so I investigated and found that the config path of the unraid template is wrong. It is mapped to /config in the container but invidious uses /invidious/config. That's the reason it's looking for postgres as host even when you specified an IP in your config file and tries to connect to the db with the default "kemal" as password simply because it uses the example config instead of your customized one.

I was literally writing up a different answer at the same time but by adding the INVIDIOUS_CONFIG_FILE variable but this is the correct answer. The mapping is indeed incorrect.

Link to comment
On 10/3/2021 at 7:34 PM, Netti93 said:

regarding invidious setup problems:
After I got past the postgres hostname problem by adding a dns entry to my pi-hole and using --dns as extra parameter for this container the logs stated that the password for my DB was incorrect. As I could login manually I had a suspicion and changed the db users password to "kemal" and that did the trick so I investigated and found that the config path of the unraid template is wrong. It is mapped to /config in the container but invidious uses /invidious/config. That's the reason it's looking for postgres as host even when you specified an IP in your config file and tries to connect to the db with the default "kemal" as password simply because it uses the example config instead of your customized one.

 

On 10/3/2021 at 11:09 PM, smalls72 said:

I was literally writing up a different answer at the same time but by adding the INVIDIOUS_CONFIG_FILE variable but this is the correct answer. The mapping is indeed incorrect.

Happy to be incorrect if others have found something.

 

The only way I could get it to work again was

 

/invidious/config  -> /mnt/user/appdata/invidious/config/

 

Is that what your referring too?

Link to comment
On 10/6/2021 at 9:33 AM, Joshndroid said:

The only way I could get it to work again was

 

/invidious/config  -> /mnt/user/appdata/invidious/config/

 

Is that what your referring too?

 

I can confirm this was my missing piece. I was able to instantly get this running through your directions using a custom network.  But I wanted to route this container through one of Binhex's VPN containers.  Once I changed the container to none and routed through the VPN, it never resolved 'postgres'.  I kept changing the config to host IPs / container IPs with no luck and was stuck for a while.  - I should have read through the posts.

 

Making the above change fixed my problem instantly. The moment I changed the path variable and restarted the container it picked up the IP and instantly loaded.  Invidious looks like it works through a reverse proxy wither or not it has the port, domain or https_only config values set.  I'm guessing it doesn't correctly link without them set, and instead would give the IP:Port/video id.  But since it looked like it worked, I assumed it was reading the config.  Knowing this, it most likely would work on bridge connections by directly pointing to the postgres container IP.

 

Anyways, thanks for putting this container together!

Edited by BiGBaLLA
  • Thanks 1
Link to comment
On 10/11/2021 at 1:18 AM, BiGBaLLA said:

 

I can confirm this was my missing piece. I was able to instantly get this running through your directions using a custom network.  But I wanted to route this container through one of Binhex's VPN containers.  Once I changed the container to none and routed through the VPN, it never resolved 'postgres'.  I kept changing the config to host IPs / container IPs with no luck and was stuck for a while.  - I should have read through the posts.

 

Making the above change fixed my problem instantly. The moment I changed the path variable and restarted the container it picked up the IP and instantly loaded.  Invidious looks like it works through a reverse proxy wither or not it has the port, domain or https_only config values set.  I'm guessing it doesn't correctly link without them set, and instead would give the IP:Port/video id.  But since it looked like it worked, I assumed it was reading the config.  Knowing this, it most likely would work on bridge connections by directly pointing to the postgres container IP.

 

Anyways, thanks for putting this container together!

 

Thank you mate for the confirmation

I'll work on updating the template shortly.

Link to comment

After the recent update to Invidious, I no longer be able to reach the Invidious instance. Seems to be caused by invidious not being able to communicate with postgres. restarting both dockers does not fix this issue...

thing is that I could use Invidious fine just yesterday, does anyone know how to fix this?

from /usr/share/crystal/src/socket/ip_socket.cr:18:5 in 'new'
from lib/pg/src/pg/connection.cr:13:23 in '->'
from /usr/share/crystal/src/primitives.cr:266:3 in 'build_resource'
from /usr/share/crystal/src/int.cr:524:11 in 'build_database'
from src/invidious.cr:38:1 in '__crystal_main'
from /usr/share/crystal/src/crystal/main.cr:110:5 in 'main'
from src/env/__libc_start_main.c:94:2 in 'libc_start_main_stage2'
Unhandled exception: (DB::ConnectionRefused)
from lib/sqlite3/src/sqlite3/connection.cr:10:11 in '->'
from /usr/share/crystal/src/primitives.cr:266:3 in 'build_resource'
from /usr/share/crystal/src/int.cr:524:11 in 'build_database'
from src/invidious.cr:38:1 in '__crystal_main'
from /usr/share/crystal/src/crystal/main.cr:110:5 in 'main'
from src/env/__libc_start_main.c:94:2 in 'libc_start_main_stage2'
Caused by: Cannot establish connection (PQ::ConnectionError)
from lib/pg/src/pq/connection.cr:34:9 in 'new'
from lib/pg/src/pg/connection.cr:13:23 in '->'
from /usr/share/crystal/src/primitives.cr:266:3 in 'build_resource'
from /usr/share/crystal/src/int.cr:524:11 in 'build_database'
from src/invidious.cr:38:1 in '__crystal_main'
from /usr/share/crystal/src/crystal/main.cr:110:5 in 'main'
from src/env/__libc_start_main.c:94:2 in 'libc_start_main_stage2'
Caused by: Error connecting to 'localhost:5432': Address not available (Socket::ConnectError)

 

Link to comment

That's really strange.  Assuming the recent change was for the config, that error message looks like it can't find postgres, but I did notice I have two configs in my template.

 

/ $ diff /invidious/config/config.yml /config/config/config.yml 

 

They're both the same for me and Invidious is running after most recent pull, but I'm not sure which is being used. I'd start by validating you have postgres on that port with the credentials and didn't modify anything outside the default.

Link to comment
2 hours ago, BiGBaLLA said:

That's really strange.  Assuming the recent change was for the config, that error message looks like it can't find postgres, but I did notice I have two configs in my template.

 

/ $ diff /invidious/config/config.yml /config/config/config.yml 

 

They're both the same for me and Invidious is running after most recent pull, but I'm not sure which is being used. I'd start by validating you have postgres on that port with the credentials and didn't modify anything outside the default.

ok, I managed to fix the issue. my config.yml was messed up for some reasons with recent pull. with that, ran 

wget -O /mnt/user/appdata/invidious/config/config.yml https://raw.githubusercontent.com/iv-org/invidious/master/config/config.example.yml

to get clean config.yml then, changing the value of

host: localhost

to actual local IP fixed the issue for me.

I do also have 2 configs appear on Container setting menu, idk if that caused the problem or not.

Link to comment
On 10/13/2021 at 9:41 PM, BiGBaLLA said:

That's really strange.  Assuming the recent change was for the config, that error message looks like it can't find postgres, but I did notice I have two configs in my template.

 

/ $ diff /invidious/config/config.yml /config/config/config.yml 

 

They're both the same for me and Invidious is running after most recent pull, but I'm not sure which is being used. I'd start by validating you have postgres on that port with the credentials and didn't modify anything outside the default.

 

On 10/14/2021 at 12:25 AM, mydisplayname said:

ok, I managed to fix the issue. my config.yml was messed up for some reasons with recent pull. with that, ran 

wget -O /mnt/user/appdata/invidious/config/config.yml https://raw.githubusercontent.com/iv-org/invidious/master/config/config.example.yml

to get clean config.yml then, changing the value of

host: localhost

to actual local IP fixed the issue for me.

I do also have 2 configs appear on Container setting menu, idk if that caused the problem or not.

 

I have learnt something else new in that it appears that if you update the template with changed mount path locations the updated template does not remove what you already had set while it adds the new one, causing what you are seeing with the dual configs. I guess this change is more helpful for new users rather than existing people.

 

I was able to delete the duplicate config path location from the edit screen, with reboot, and invidious was working as it was before.

This is mine for reference.

 

1660148209_Screenshotfrom2021-10-2014-25-34.thumb.png.2f119cfcbfed2c27d47892f18195012c.png

 

Hopefully we are over the hump of some of these changes... I have yet to make any changes to my own setup other than what others are having issue with and suggesting is the culprit. I have then been getting my instance back working with their suggestions and updating the template to reflect it. Apart from that its been solid as a rock for me, I totally forget about it and just update it every few days.

Link to comment

I'm seeing the following in my logs.  Latest update.  It was once working fine via reverse proxy so I don't want to change anything on that end, it seemed to stop working with one of the container updates.

 

Unhandled exception: could not find expected ':' at line 100, column 1, while scanning a simple key at line 98, column 1 (YAML::ParseException)

Link to comment
On 10/22/2021 at 11:32 AM, darkrom said:

I'm seeing the following in my logs.  Latest update.  It was once working fine via reverse proxy so I don't want to change anything on that end, it seemed to stop working with one of the container updates.

 

Unhandled exception: could not find expected ':' at line 100, column 1, while scanning a simple key at line 98, column 1 (YAML::ParseException)

 

check your config.yaml and that it is in the correct location.

The changes should enable the config.yaml to actually function better and it may be now picking up errors that were originally not picked up... You can try pasting the config.yml into an online yaml testing site and see if there are any errors if you are unsure.

Link to comment
On 10/25/2021 at 12:57 AM, Joshndroid said:

 

check your config.yaml and that it is in the correct location.

The changes should enable the config.yaml to actually function better and it may be now picking up errors that were originally not picked up... You can try pasting the config.yml into an online yaml testing site and see if there are any errors if you are unsure.

 

 

It is the same config.yaml that was working before one of the updates and then it stopped. 

It is located in   /mnt/user/appdata/invidious/config/config.yml

I'd love to get this working again.  I'll try to find a testing site.

Link to comment

Regarding Baikal ..

I just installed from ckulka/baikal:nginx.

Installation OK.

While doing the config I get following error:

 

Validation error

The FOLDER containing the DB file is not writable, and it has to.
Please give write permissions on folder /var/www/baikal/Specific/db

 

I only have empty baikal/Specific/ folder, no db.

Any ideas how to solve the problem?

 

Regards

Edited by matty2k
Link to comment
  • 2 weeks later...
On 11/3/2021 at 7:57 PM, matty2k said:

Regarding Baikal ..

I just installed from ckulka/baikal:nginx.

Installation OK.

While doing the config I get following error:

 

Validation error

The FOLDER containing the DB file is not writable, and it has to.
Please give write permissions on folder /var/www/baikal/Specific/db

 

I only have empty baikal/Specific/ folder, no db.

Any ideas how to solve the problem?

 

Regards

 

I noticed you have an 'S' in your location

My folder location is '/mnt/user/appdata/baikal/specific'

I know that baikal is looking for Specific but I kept everything as lower case in the template.

I also have mine set up to use an external MySQL database, maria db, so my /specific only have 2 files 'INSTALL_DISABLED' and .htaccess

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.