How to prevent VSCode scan installed applications? - visual-studio-code

VSCode shows the following message every time I start it:
You have Docker installed on your system. Do you want to install the recommended extensions for it?
The message appears even if Docker is not running at the time when I start VSCode. So I think VSCode scanned the file system to look for what applications I have installed.
Is there any way to opt out from this? This feels like an invasion of my privacy, I do not want Microsoft know what apps I installed. Turning off telemetry does not help, the message still appears.

Related

Developing Flutter with VSCode and WSL2

Since I mostly develop Web, using nginx, PHP and MySQL, I have ported my WebDev-environment entirely to WSL2.
Since performance is very important, all my web-related projects reside on the WSL2-vhdx file /home/user/Projects/Web. In WSL2 I've installed all my necessary tools for a nice and neat Linux-like experience, Docker, GIT, etc.. This combined with VSCode remote integration works very well.
Now, I'm digging into building Flutter-Apps, and my Flutter-environment is installed on the Windows side. My Flutter-related projects reside on D:\Projects\Flutter which is a partition, and NOT USED in WSL2 in any way. Building Flutter-apps with flutter-windows-sdk and VSCode works neatly.
But, the problem is: Now I've my project files scattered all across my computer. Web-stuff in a WSL2-vhdx-file and Flutter-stuff on the D-partition.
Is there a way to build flutter-apps with Flutter, while having the project-files stored on a WSL2-vhdx-file, in combination with VSCode-remote and an Android-emulator?
I tried creating a test Flutter-project on the \\wsl$ network mount, which didn't work.
Moving my web-related project files to the D:\ partition of Windows is no option, since the I/O mounts in WSL2 are extremely slow.
I got it working, reliably with adb connect 192.168.xxx
For anyone interested, see my full blog post here: https://dnmc.in/2021/01/25/setting-up-flutter-natively-with-wsl2-vs-code-hot-reload/
Is there a way to build flutter-apps with Flutter, while having the project-files stored on a WSL2-vhdx-file, in combination with VSCode-remote and an Android-emulator?
I'm assuming (based on the mention of VS Code Remoting) that you want to run the extension in WSL. I haven't tried that specifically, but I have run Flutter inside WSL and also connected a VS Code Remoting session to an Android emulator in the cloud, so I would expect this to work.
You'll need to make sure you set up the Flutter SDK inside WSL (so you can run flutter commands inside WSL - it should be the Linux version of the Flutter SDK and not the Windows one if you're using the zip).
To have your emulator show up in flutter devices from inside WSL, you will likely need to run adb tcpip 5555 from the Windows side (this means you need an Android SDK in Windows) - this will tell your phone to listen on TCP port 5555. Then you'll need to run adb connect [phone ip]:5555 from inside WSL (this means you'll need an Android SDK in Linux). If all goes well, the phone should then show up in adb devices and also be picked up by the device selector in VS Code.
I tried creating a test Flutter-project on the \wsl$ network mount, which didn't work.
It's not clear what went wrong here, though my first guess would be that maybe the UNC path isn't supported - if you map a drive letter to it does it make a difference?
While this isn't an officially supported setup, feel free to raise issues in the Dart-Code repository on GitHub with any issues you have. It's not a priority, but I would like for VS Code Remoting (including WSL and Docker) to generally work for Dart and Flutter dev.
Anytime you're crossing/sharing the file-system boundary from windows to wsl you're paying a massive cost in speed/time.
With the setup you've described I'd consider trying to self-host the browser based VSCode.dev inside wsl - checkout details instructions here: https://medium.com/geekculture/3-steps-to-code-from-anywhere-45401247f479
Personally I've settled on running VSCode and docker inside a Linux VM on Windows, and have a 96% time saving in things like running up a server and watching code for changes making this setup my preferred way now.
The standardisation of devcontainer.json and being able to use github codespaces if you're away from your normal dev machine make this whole setup a pleasure to use.
see https://stackoverflow.com/a/72787362/183005 for detailed timing comparison and setup details

Why won't my Click-Once app install?

I have an app I wrote in WPF, which I publish by right-clicking on the project in Visual Studio 2013. This produces a set of files which I zip up and make available for users to download and install.
This has been working fine for a while, but today a user told me that he can't install the latest version. It gets as far as the install dialog, then just hangs there. If you kill the install, the app doesn't show up in the list of installed programs, but in Task Manager, I can see that the app itself is actually running, but has three instances running. I can't kill any of these, and have to restart the machine.
He's using Windows 7 Home Premium if it makes any difference.
I've tried uninstalling the old version and restarting his machine several times, but it doesn't help. The app installed fine on my machine, and as far as I know, nothing has changed on his machine.
Any ideas what's gone wrong? Don't know if I missed out any important details. If so, please let me know and I'll fill them in.
In case it helps anyone, it turned out to be to do with the certificate. I thought I was using one, but his machine seemed to think I wasn't. We had to turn off UAC and then right-click the exe to unblock it. That got him going.
I'm fed up with Click-once. it's been nothing but trouble. I've been playing with InnoSetup, which works really well and doesn't have all these problems. It's also pretty easy to make the app update itself automatically without driving the users mad.
Hope this helps someone.

How Do I debug in server side code?

I have download & install vscode on my ubuntu 14.04 lts box, and It's working fine. I have created one app in node.js, and want to debug it. When I put debug point in my app.js file, and hit f5 (run), and seems it's working. But How to debug server side and client side code? Do you have any video tutorial in detail, so new user can understand it.
Thanks
VSCode only supports node and mono (on linux and mac) debugging atm.
Client side debugging is not currently supported, you can vote for it here: https://visualstudio.uservoice.com/forums/293070-visual-studio-code?filter=hot&page=1
We might produce video tutorials for new users soon.
In the meantime you can look at the docs
https://code.visualstudio.com/Docs/debugging
The VSCode debugger does not support to debug clients (browsers) because browsers already have built-in debuggers (and other dev tools) that are much better integrated than VSCode can ever be.
So please use VSCode for debugging your node server and for debugging the client at the same time open the development tools of your browser.

Why is VPN connection on Mac OS X causing Eclipse to lockup during 'edit launch config' operation?

I have been using Eclipse on Mac OS X from home over a VPN to develop GWT and perl code in a local workspace for my employer. Recently a repeatable and severe lockup began occurring whenever I tried to edit Debug (or Run) launch configurations. I got the spinning beachball of death (SBOD) and, if I waited long enough (10-15 minutes) it would eventually stop and I could at least close the dialog.
I tried numerous things until a coworker suggested trying it with VPN turned off. To my surprise (and somewhat delight) it began behaving normally in the above scenario. I have been using Eclipse in this manner for about a year with no problems so naturally I am racking my brain trying to think of recent changes to VPN and/or to my split tunnel script (euphemistically called 'multihome') that could account for this abnormal behavior. This lockup occurs with or without the split tunnel.
I should also point out that the "initializing Java tooling" progress status ALWAYS occurs when starting eclipse and takes about a minute to complete with VPN connection. Normal (<2 seconds) without.
So, I'm starting to learn how to use wireshark and possibly will look into using packetlogger as well in an attempt to find out more about this strange issue.
Anyone have a clue as to what might be causing this?
This is a hard one to answer. Short answer is "I don't know".
However, I did find out that, due to the recently updated JVM, done as part of the Mac OS X update, Eclipse IDE lost its ability to find the src.jar file for the JRE. As a result, it appears that Eclipse, in various places in the code, searched for this file and when not found attempted to find it via the network. When VPN was turned on, perhaps this exacerbated the problem.
This was solved by fixing the Installed JRE configuration of Eclipse (see JDK on OSX 10.7 Lion).

Anyone have issues with Eclipse over Remote Desktop Connection?

I have a very strange issue that I'm hoping someone can help me with. I have various installations of Eclipse on my development machine at work. The one I primarily use is Weblogic WorkSpace Studio 10.2. This installation, along with a few Pulse installations I have set up works fine when I'm logged into my computer physically.
However, when I try to log into the computer using Microsoft's Remote Desktop Connection utility I get an error stating: "Could not create Java virtual machine." and then I get the lovely Eclipse error box which I personally can gather almost nothing from.
Even if you don't have the solution, any help would be greatly appreciated.
Justin
What ended up working for me was the memory settings for the JVM. Apparently the remote desktop connection, or some other setting in Windows, blocks off a fairly large amount of space. By reducing the heap size allocation for the JVM during Eclipse and server start-up I was able to get this working. As a side note, I had PLENTY of space that windows could have used, so I don't think blankly adding more memory would necessarily solve the issue. If you find another solution, please let me know.
• We came across an issue when user RDC’s to a remote system where the OS is Windows 10 and has a running Eclipse instance, the Eclipse instance terminates
• Eclipse is one of the IDE’s for Java
• The issue is because of Windows 10 Exploit protection
• Pre-requisite: You will need Administrative permissions for executing the below
• Navigate to Settings -> Update & Security -> Windows Security -> App & Browser Control -> Exploit Protection Settings
• Add the program to exclude as below
P.s. As of Window 10 1909 MS security advisory mentions we can disable some exploit protections by default.
Perhaps it is permission related. take a look at similar issue that symantec has:
http://service1.symantec.com/support/ent-security.nsf/854fa02b4f5013678825731a007d06af/8ea1593f1d1fcee68025759a003d8403?OpenDocument
Try to see if you have same patches installed that causes the security issue. Also refer to application log to see if there is a more specific error. Good luck :)
I think issue happens due to Windows, not Eclipse nor JVM. There is still open Bug report on the Eclipse side and one of the comments state that Microsoft is working on the issue.
I have tried Windows Remote Desktop-ing into my dev machine at work (which had only one version of eclipse installed on it). I had no troubles.
Is it possible that your problems stem from multiple versions of eclipse running at the same time?
Also, have you tried a fresh install of eclipse on your dev machine?
If the above two suggestions don't work, then the only thing that I can think of is what Mohammad said: you might need to check your permissions.
I would check the system log if I were you: Start > run > eventvwr
The first thing to look at is the .log file which is in your eclipse's metadata folder (found in your workspace at $WORKSPACE_ROOT/.metadata/.log). If you post the stack trace that it generates upon initialization, we can give a definitive answer.
I am now experiencing this in Eclipse (the Oxygen release and Java 1.8.0_181). I previously had the same problem with another Java-based program (Oxygen XML/XSL editor - the product name is coincidentally the same as the Eclipse version). Last year the Oxygen support team answered that it may be a known problem in Java.
Even without seeing a crash report, considering your sequence of
events, this seems like a known common cause of crash for the Java
runtime. Keeping Oxygen/Java running for a long time, until the screen
or video card enters sleep then connecting/disconnecting
screens/projectors or connecting/disconnecting RDP can trigger a crash
in the Java runtime. We keep updating the Java runtime (JRE) with each
new version of Oxygen, but so far the issue has not been resolved in
newer versions of the JRE.
e.g. Java VM logged issue:
https://bugs.openjdk.java.net/browse/JDK-8153389