live-server | A simple development http server with live reload capability

I have a master slave setup with 1 master and 2 slaves. When I kill the master, one of the slave tries to become master but fails with following exception:

Goodnight all.

When I try to install a package I get the error you can see below and nothing installs.

I'm trying to have live-server auto reload the code once I save my change.

I've installed live-server globally and created a folder test and put 3 files in there

I last used it on wednesday last week when I worked and it worked perfectly fine with no problems. Now that I came back today and wanted to start it with live-server it showed symbolds in the console and even waiting for 5 minutes nothing else happaned.

The weird stuff which is showing up

It tried to use powershell instead of the command prompt. At first it told me that scripts are disabled so I searched that, enabled scripts and the exact same thing then happaned as in the command prompt.

I didn't find anything about this weird problem. I even installed the oldest version of it but nothing changed.

For context, this issue doesn't happen on my laptop but it happens on my desktop at home My desktop and office pc both are a few years old in terms of hardware but couldn't be any more different besides windows 10 pro.

So, I'm trying to follow along this tutorial and right off the bat I can't get it to work. First off, these are my files:

root/frontend/index.html

On my Windows machine I set up a local test cluster of 2 brokers (version 2.19.0): 1 master, 1 slave. The ha-policy is replication and cluster communicating goes over JGroups.

master broker.xml:

I have the problem that when I use a custom http agent, sockets are not released when an error occurs during the piping of the response.

It seems that Node.js isn't doing this on it's own. Thats why I tried to release it myself by destroying the response and the underlying socket but it didn't work. Nothing I tried worked expect destroying everything with agent.destroy(). This obviously can't be the solution because like this properly running sockets are getting destroyed.

That leads me to my question. What is the proper way to release a socket?

Following an example to reproduce it.

• Run the script below with node index.js
• Open http://localhost:3000/
• Cancel download

• You will see the socket remains occupied forever.

If you call http://localhost:3000/small you see the behaviour I would expect to happen. Which is that the socket is getting released.

I have no idea what I did wrong. I'm trying to make right configuration of ActiveMQ Artemis HA cluster on 2 docker containers, but I don't see in logs anything about "backup server" and so on. It seems to not work. I tried to make a simple Java client for the cluster, but it doesn't connect to it. Please don't write "have you read a documentation" - because yes, I've read it. From my perspective it seems like there is no connection between master and slave, but I don't know why. I've configured docker environment with data:

Dockerfile:

I have 3 pair master/slave Artemis HA cluster setup with replication. They are running on 6 different machines and each of them using their own self signed certificate where CN matches the hostname where the brokers are running. It was new in artemis 2.18.0.

broker.xml and bootstrap.xml with sslEnabled settings looks like this: