Multi user support

[BossHogg97]

Hi, i'm working with trame and pyVista for visualize vtu files. Everythink is working fine with a single user. I'd like to implement a mechanism that allow different users to work on diffent vtu files in a separete state. At the moment on my main.py (entry point) i istanciate a viewer class. How can i achieve a multi user setup?

Followin main.py:

import argparse
import sys
from loguru import logger
import os

from ui.viewer import TrameViewer

def main():
    
    args = parse_arguments()
    setup_logging(args.debug)

    # Initialize viewer app
    viewer = TrameViewer()
    
    print("=" * 60)
    print("VTK Viewer started!")
    print(f"Trame UI: http://{args.host}:{args.port}")
    print("=" * 60)
    
    try:
        # Start Trame (blocks main thread)
        viewer.start(port=args.port)
    except KeyboardInterrupt:
        logger.info("🛑 Shutting down application...")
    except Exception as e:
        logger.error(f"❌ Trame server error: {e}")


# Other utilities functions

if __name__ == "__main__":
    main()

Following the init function of viewer.py class

# PyVista
import pyvista as pv
from pyvista.trame.ui import plotter_ui
from pyvista import examples

# Trame Imports
from trame.app import get_server
from trame.app.file_upload import ClientFile
from trame.ui.vuetify3 import SinglePageWithDrawerLayout
from trame.widgets import vuetify3, trame  # This import provide the access to trame's widget for vtk and vuetify
from trame.ui.vuetify3 import SinglePageWithDrawerLayout

# Core Imports
from config.enums import ColorMap, RenderMode
from config.settings import config

pv.OFF_SCREEN = True

class TrameViewer:
    """Interfaccia grafica Trame + PyVista per visualizzare file VTU con controllo avanzato."""

    def __init__(self):
        # --- SERVER TRAME ---
        self.server = get_server(client_type="vue3")
        self.state = self.server.state
        self.ctrl = self.server.controller
        
        # --- CONFIGS ---
        self.config = config

        # --- PLOTTER & PIPELINE ---
        self.plotter = pv.Plotter(off_screen=True, notebook=False, window_size=(800, 600))
        self.mesh = examples.load_random_hills()
        self.plotter.add_mesh(self.mesh)

        # --- FILE-UPLOAD ---
        self.state.file_exchange = None
        self.upload_vtu_file()

        # --- Initialize state for mesh card UI ---
        self.state.setdefault("active_ui", None)
        self.state.setdefault("mesh_color_selection", 0)
        self.state.setdefault("mesh_representation", RenderMode.SURFACE)
        self.state.setdefault("color_map", ColorMap.COOL_WARM)
        self.state.setdefault("mesh_opacity", config.default_opacity)
        self.state.setdefault("show_edges", True)

        ## --- Initialize state for contour card UI ---
        self.state.setdefault("num_contours", 10)
        self.state.setdefault("contours_opacity", 1)


        self.state.mesh_options = []

        # --- Create UI ---
        self.create_ui()

        # --- Setup Callbacks ---
        self.setupCallbacks()
        
        # Other code that manage callbacks and define UI
        

         # ----------------------------------------------------------
         # SERVER START
         # ----------------------------------------------------------
         def start(self, port=3015):
             """Start trame (block main process)"""
             self.server.start(port=port)

Thanks :)

[clabnet]

Interesting question, I'have not found a solution for the same use case.
Many posts says about a "launcher" parameter but is there not documentation about it.

[jourdain]

The multi user setup can be achieved with the docker bundling. See those examples for reference.

If you want to configure each session differently based on some URL parameter, you need to rely on the CLI and template some of your parameters with the ones you will get from the URL inside the apps.yml like we do with host and port.

HTH

[BossHogg97]

Thanks for the prompt reply. I updated my docker file, now using:

FROM kitware/trame:uv-glvnd

ENV TRAME_PYTHON=3.13

RUN apt-get update \
    && apt-get install -y \
    libxrender1 \
    && rm -rf /var/lib/apt/lists/*

COPY --chown=trame-user:trame-user . /deploy

RUN /opt/trame/entrypoint.sh build

And added setup folder with apps.yml and requirements.txt. Following requirements.txt

trame
trame-vtk
trame-components
trame-server
trame-vuetify
vtk
loguru
python-dotenv
pyvista

Now i can build the docker image correctly but when i run docker run -it --rm -p 3015:80 trame-app the UI still on loading and after several seconds i've got these errors on console:

* Restarting Apache httpd web server apache2                                                                                                                                                                                            AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.17.0.2. Set the 'ServerName' directive globally to suppress this message
                                                                                                                                                                                                                                  [ OK ]
Starting server...
Starting the wslink launcher at
======== Running on http://0.0.0.0:9000 ========
(Press CTRL+C to quit)
Error handling request from 127.0.0.1
Traceback (most recent call last):
  File "/deploy/server/venv/lib/python3.13/site-packages/aiohttp/web_protocol.py", line 510, in _handle_request
    resp = await request_handler(request)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/deploy/server/venv/lib/python3.13/site-packages/aiohttp/web_app.py", line 569, in _handle
    return await handler(request)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "/deploy/server/venv/lib/python3.13/site-packages/wslink/backends/aiohttp/launcher.py", line 97, in handle_post
    return await self._waitForReady(session, request)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/deploy/server/venv/lib/python3.13/site-packages/wslink/backends/aiohttp/launcher.py", line 109, in _waitForReady
    if self.process_manager.isReady(session, count):
       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^
  File "/deploy/server/venv/lib/python3.13/site-packages/wslink/launcher.py", line 480, in isReady
    if not self.isRunning(id) and count < 60:
           ~~~~~~~~~~~~~~^^^^
  File "/deploy/server/venv/lib/python3.13/site-packages/wslink/launcher.py", line 468, in isRunning
    return self.processes[id].poll() is None
           ~~~~~~~~~~~~~~^^^^
KeyError: '82e6704c-b4d8-11f0-b3d1-672a89746aa3'

Thanks

[jourdain]

Do the docker examples from the trame repo works for you?

[jourdain]

To see why your app failed, look at the logs in /deploy/server/logs/launcher/*.txt where * in your case would be 82e6704c-b4d8-11f0-b3d1-672a89746aa3

[BossHogg97]

Do the docker examples from the trame repo works for you?

I'll try as soon as possible, i tried directly on my repo

[BossHogg97]

The error inside is the following:

usage: main.py [-h] [-f FILE] [-r] [-p PORT] [--host HOST] [--debug]
main.py: error: unrecognized arguments: --authKey F9MvpkXptByy5fm3 --data ${data}

[BossHogg97]

I commented the following lines and is working

# - --authKey
# - ${secret}
# - --data
# - ${data}

[jourdain]

Ah that is because you are messing up with the args a trame app can expect...

You should be using:

server.cli.add_argument(...)
server.cli.parse_known_args()

[jourdain]

https://github.com/Kitware/trame/blob/master/examples/00_howdoi/cli.py

[jourdain]

--data ${data} is not expected on the trame side though.

[BossHogg97]

Thank you so much for your fast support!