Sharing Panel Dashboards

Panel
1

Servable works but for myself. Is there any way I can share it internally in my organization who have no anaconda and have no clue about all these. They are the end users who can interact with the dashboard and take actions.

Currently I am exporting to html which makes it static.

Thanks,
Anand

2

Hi @AnandSamant,

Yes, how you do it may depend on the IT infrastructure in place. In short I have a dedicated machine, this machine serves my simple apps and then users can connect to them by pointing their web browser at dedicated machine name & port. With my small user base this works really quite well within the company cloud.

https://panel.holoviz.org/user_guide/Server_Configuration.html

I more or less run this command from the prompt

panel serve app1.py app2.ipynb

The latest version of panel supports a new method where you may not even require a server running but I haven’t looked at this further but seems very promising!

I can go through what I do in more detail but will be later as just started tattie holidays here.

Anyway hope some of the above will point you in right direction that it is all possible thanks Carl

1 Like
3

Thank you so much Carl. I can certainly wait for you to come back from your holidays. I have used panel serve but works just for me. I do have a separate PC which will be running 24*7 (my server) and I just need help to understand how your users connect to your server (like what do they enter on the web browser). I have tried replacing localhost with ip address but it did not work.

Happy Holidays!

4

Hi @AnandSamant,

Simply I enter

http://ipaddress:port
e.g. http://10.146.4.97:5006

port in my case is 5006 which I think is the default port, you may need to allow access of this port through the computer acting as the server firewall.

on my windows machine the firewall looks something like this

image
image627Ă—558 25.3 KB

Hope it helps, Carl.

2 Likes
5

maybe you need to use

panel serve app.py --allow-websocket-origin="*" --address=0.0.0.0

to serve the app. The app will be in the location that Carl mentioned, http://ipaddress:port for the users inside your private network

2 Likes
6

I did both of this…Final solution for me is as below

  1. Made the template servable
  2. Created this Firewall setup suggested by Carl
  3. Went to my command prompt and entered panel serve filename.py or ipynb follow by --allow-websocket-origin = ip address of the server I am running this ipynb file from followed by :5006

panel serve Dash1.ipynb Dash2.ipynb --allow-websocket-origin=x.x.x.x:5006

1 Like