pludov / mobindi Goto Github PK
View Code? Open in Web Editor NEWControl astronomy software from a mobile browser
License: GNU General Public License v3.0
Control astronomy software from a mobile browser
License: GNU General Public License v3.0
Hello,
Not sur if real problem or not since it's in a "Next release" paragraph, but I have trouble launching mobindi in astroberry
How i launch :
astroberry@astroberry:~/mobindi $ cat mobindi.conf
LOGDIR='/var/log/mobindi'
PORT=8081
astroberry@astroberry:~/mobindi $ ./install.sh
Settings:
* Log directory: /var/log/mobindi
System ready
astroberry@astroberry:~/mobindi $ ./startup.sh
Nginx reloaded
> [email protected] start
> node dist/app.js
astroberry@astroberry:~/mobindi $
Logs
2021-07-03T09:43:01.087Z info [Log.js] Starting
2021-07-03T09:43:01.608Z error [app.js] Got express error listen EADDRINUSE: address already in use :::8080 Error: listen EADDRINUSE: address already in use :::8080
at Server.setupListenHandle [as _listen2] (node:net:1305:16)
at listenInCluster (node:net:1353:12)
at Server.listen (node:net:1440:7)
at init (/home/astroberry/mobindi/app.ts:274:12)
at Immediate.<anonymous> (/home/astroberry/mobindi/app.ts:319:9)
at processImmediate (node:internal/timers:464:21) {
code: 'EADDRINUSE',
errno: -98,
syscall: 'listen',
address: '::',
port: 8080
}
2021-07-03T09:43:01.709Z info [Log.js] Process terminating with code 0
I've had some more success with exporting, then install, then launch instead of using mobindi.con
export PORT=8082
I can then acces UI, and even take pictures, but then won't come back to UI
>>capture<<
;)
Hello.
Everything works fine.
I am on Raspberry Pi 4B 4gb of ram.
Unfortunately it is impossible to operate the automatic start.
I try via a service or via the folder automatic start but impossible to start mobindi.
All failed.
for autostartup folder :
echo -e '#!/bin/bash'"\ncd /home/$USER/bin/mobindi\n/home/$USER/bin/mobindi/startup.sh &\n" > /home/$USER/bin/mobindi_up.sh
chmod +x /home/$USER/bin/mobindi_up.sh
echo -e "[Desktop Entry]
Encoding=UTF-8
Version=1.1.0
Type=Application
Name=UI_Indi
Comment=Contrôle de Indi
Exec=/home/$USER/bin/mobindi_up.sh
StartupNotify=false
Terminal=false
Hidden=false" > /home/$USER/.config/autostart/mobindi.desktop
chmod +x /home/$USER/.config/autostart/mobindi.desktop
for service :
dir=/etc/systemd/system #mettre le chemin du service en memoire
MOI=${USER}
sudo updatedb # mettre à jour la base d'indexation
dir_mobindi="/home/$USER/bin/mobindi/startup.sh" #récupérer le chemin du binaire browsepy
echo -e '[UNIT]\nDescription=MobIndi\nAfter=multi-user.target\n\n[Service]\nType=idle\nUser='${MOI}'\nExecStart='${dir_mobindi}'\nRestart=Always\nRestartSec=5\n[Install]\nWantedBy=multi-user.target' >> /tmp/mobindi.service
sudo cp /tmp/mobindi.service ${dir}/
sudo rm /tmp/mobindi.service
sudo chmod 644 ${dir}/mobindi.service
sudo systemctl daemon-reload
sudo systemctl enable mobindi.service
sudo systemctl start mobindi.service
Hi,
Congratulations for this great soft!
I'm not sure to understand where are stored the picture taken during the session. Is it stored on the Pi SD or on the phone ?
Have you already considered to add support for the Alpaca solution of ASCOM to control the mount ?
I'm currently trying to create a small very portable GT mount and your soft is exactly what I wanted to have to control the mount.
I'm open to help for this implementation
Display the locked star on frames when PHD display is available
Depends on issue #26
Current code assume fits files are immutables.
Gives a way to invalidate a file (killing associated preview process/cache/...)
This is especially required for continuous shoot.
This will require collaboration from fitsprocessor:
* add a pre-trash notification to fitsprocessor
* when no new request is flying, drop the file & the associated cache content
* refuse further access to resource
Display the frames acquired by PHD2 from the guiding UI
For perf reason, this must not be displayed by default here
When a sequence include dark frames, mobindi should ask the user to cover scope.
Until acknowledged, the sequence must be put in confirm state. A special dialog must appear.
A system notification must be pushed if no client is active to display the dialog.
When an indi driver is shutdown , the ui is not updates. The properties of the device stay visible with their last value.
hello.
I'm testing the nafabox with the new 22.04LTS but unfortunately I can't install mobindi.
Unable to pass the installation step.
Attached is the output of ./install.sh
thanks for help
Allow setting the range and settle time in sequence editor
Autofocus UI should:
Seems not to build correct, i get a lot of errors and warnings.
There is no mobindi/dist folder at all.
Tried on Raspberry pi 3 with astroberry-server image (ubuntu16.04) an on my laptop with ubuntu 18.04 x64
mobindi.txt
I attach the complete console output of the build process.
Please help
The tools already exist in examples, they just have to be installed/configured by default.
Also there is a need for a UI to display details of error messages when they
occurs (typically, check if the current unix user is not allowed to
shutdown)
Add a notification for low disk space condition.
The notification must appear in-app then in system notification. This is a
permanent notification (expected to stay until acked by user, and can disappear by itself)
There must be a place where the thresold is set (a configuration page). click on
the notification should display / open this page.
The config page must display the available disk space on storage partition.
Add an option on indi device configuration to declare a filterwheel as "manual".
When the sequence require a change on such a filterwheel, pause the sequence and ask the user to change the filterwheel.
A push notification could also be emited
There is no ui to report messages from indi drivers
Can be usefull to adapt the the search range
Forcing re-calibration of PHD is not possible currently.
Maybe expose it as long press on the button guide button
Hello.
I switched to 20.04LTS for the nafabox, on the other hand I can no longer launch Mobindi at startup.
Each time you restart, the service that starts mobindi does not start. I have to start startup.sh by hand. Then CTRL + C. Then there the service wants this launch.
The service :
`[Unit]
Description=MobIndi
After=multi-user.target
[Service]
Type=idle
User=nafabox
ExecStart=/home/nafabox/bin/mobindi/startup.sh
Restart=Always
RestartSec=5
[Install]
WantedBy=multi-user.target`
Error message :
Started MobIndi. mobindi.service: Main process exited, code=exited, status=1/FAILURE mobindi.service: Failed with result 'exit-code' /etc/systemd/system/mobindi.service:9: Failed to parse service restart specifier, ignoring: Always
Originally posted by @dragonlost in #31 (comment)
From image views, display the histogram with a predefined precision.
The histogram computation should occur on client side to allow further
interaction (like zooming, getting the value at a given point, setting
min/max level, ...)
Reporting important event to the user. Such events could be:
Running in background (while screen is off) will probably require either going PWA or native
Current integration with travis does not cover C++ unit tests.
Back button currently leave mobindi.
Add support for in app navigation using URL params/dash (React Router)
This will allow for back forth and reacting to notification
When user switch camera in the camera tab, recall the last parameters used for the newly selected camera.
This will especially be important for parameter that are specific for a camera, like gain, filterwheel, ccd temperature, ...
When switching to landscape orientation, the UI is currently displaying very big text.
The UI should adapt accordingly, possibly with multi-column layout
UI option in camera tab, to take images continuously until interrupted or used move away from camera tab.
Need to trash images after it has been used.
PHD default to the last exposure duration set by default.
Add a control in the UI tab for PHD to view/set exposure duration of PHD frames.
Add a context menu to PHD display, to select guide star
Depends on #26
hi @pludov
their is no doubt that it is an awsome project but the project does not have any logo.As a designer i want to provide a design for this project,I just need a permission so that my work does not go in vain.its free for open source projct
thanks
When configured, the camera control panel will display a filterwheel setting, with the names of the filter.
The setting must also be present on the Sequence editor.
The name of the filter should be included in the fits file name.
Some configuration may require disabling guiding during filterwheel change (CCD with integrated guide cam)
They are reset on each restart
HTTPS is now required to have access to most HTML5 advanced feature (location, pwa, ...).
This normally involves having a public DNS entry and generating a certificate for it. But that is not really a good solution for an appliance that gets its ip from a phone. Also, asking user to own a domain name is problematic
All in all SSL in the LAN context is not widely supported and will require hacky solution.
As an illustration of the problematic: https://blog.filippo.io/how-plex-is-doing-https-for-all-its-users/
In the simplest form, the app could be served by a self-signed certificate for its private ip. A new certificate needs to be generated on each ip change. (without restarting the session !)
Alternative is to go native (like cordova, electron)
The current implementation just add sync points to the model.
This becomes inefficient when lots of points have been added.
=> force the driver in using a simple sync model
focal (20.04) is not support for node_8_.x
focal is supported by the version node_10_.x ( Dubnium)
Indi drivers configuration
Indi drivers are currently started according to a json file.
This feature would provide a UI to select/deselect driver.
The UI could be located in the INDI panel, by adding a button called
"configure drivers". It would then displays the content of the json file and make
it modifiable
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.