Improvements needed in ConnectWise Configurations Page

A Rant

ConnectWise is powerful, there's no doubt there. It is even more powerful when it integrates with Labtech. Unfortunately, the user interfaces of both programs leave so much to be desired that it is really hard to be efficient on a daily basis.

The problem is that ConnectWise configurations are, after two years, still very immature from a usability standpoint.

A technician may work on hundreds of tickets each day. In theory, having computer information attached to that ticket would actually make that technician much more efficient. If you've been to a ConnectWise sales session, you know that this is the pitch you're given.

Unfortunately, in my opinion, ConnectWise still misses the mark in this area. Managing a computer from ConnectWise actually takes longer than clicking into Labtech and searching for the computer by the same name.

The Problem

I'll try to demonstrate this by counting the number of clicks it takes to manage a server via RDP. For the purposes of this demonstration I'll assume the technician already has the ticket opened.

1. See that little "view" link? That's click #1.
   
2. Now we get a bunch of information, which although potentially useful, mainly just takes up space. I am really just interested in the "Manage" button which will bring up the computer screen in Labtech. That's click #2.
   
3. Ready for things to get a bit ridiculous? We're about to have three clicks all in one step. Here we are clicking Redirectors > Network redirectors > Remote Desktop. Ok, so now we're at a whopping 6 clicks to just get a remote connection screen up.
   

Understandably, 6 clicks is small in the big scheme of things and may seem nitpick-y, but multiply that by thousands of tickets.  

I respect the power of ConnectWise, but am always floored by the shear volume of self congratulation at their user events.  They have a ton of work to do in the usability area and really haven't made any investments in this area in over two years.

A short term solution

Here is a mock up of a theoretical UI that would make me much more efficient as a technician.:

Node.js Applications as a Windows Service

Normally I try not to repost stories, but this came in so handy that I have to share it.

I was working on a getting Node.js to play nice with MongoDB. This is my first time working with Mongo so I cloned MongoClikker in an effort to bring a little gui to this text-based party. I eventually realized I wanted MongoClikker to be running at all times on my dev box (which is running Windows). Rather than having a command prompt at all times, I was able to use the instructions on Tatham Oddie's post to make MongoClikker run as a service thanks to the help of NSSM.

I was able to make this work using the following commands:

nssm.exe install mongoclikker "C:\Program Files\nodejs\node.exe" "C:\code\mongoclikker\app.js"
net start mongoclikker

So far it is working wonders!

Nginx, Comet, and Symfony2. Fun times.

What have I done?

I've officially found a brand new config that I am testing out. This time it is ngninx with the http push module to accomplish Comet pushes without the need for a full Bayeux implementation, php-fpm using a unix sock, and Symfony2 for the MVC legwork. I am looking at setting up HAProxy on the front end to do load balancing if possible in the future.

So far it has been very painless with a few caveats. I'll be interested to see how the performance of the server can handle the repeated requests. At this point I have the push module running on static html pages. I will be curious to see if php is able to take the requests as fast as I can throw them down the open connection.  Initially I'll be connecting to just MySQL dbs and maybe a SQL database occasionally. At some point I'd like to see if I can integrate with Amazon RDS. I am pretty psyched and can think of so many projects this combo could be used for.