My main node-red (3.0) installation at work is on a CentOs VM. I am using node-red-contrib-mssql-plus v0.7.3 to access some MSSQL DBs and I have no problems.
I need to make some installs on a few PCs that will serve just as connected HMIs. I just made a fresh install on a test machine to make sure it works before deployment to the final machine. I used a windows VM, and just did a fresh node.js and node-red install, with the same versions I'm running on the other machine.
I am accessing both VMs from my laptop, within the local network. However, when I want to configure the base nodes for the DBs, there are substantial differences on what they show:
This is on the CentOS machine vs the Windows machine, side by side:
Actually, it was the mssql-plus one. Same version.
I think I figured it out, it might have had something to do with the installation. Whenever I installed from the palette manager, this would happen and options would be missing.
I deleted everything, stopped node-red, and then installed the package using npm, from a powershell window running as admin.
Installation was successful, and when I restarted node-red everything was as expected.
I had the plus version. Exactly the same as in my other environment.
It was a clean install, so I only had an inject node, the mssql and a debug node. I deleted the mssql-plus package also from the palette manager after deleting the node and configuration node.
I opened an admin powershell and run the npm command as specified on the node description, and it worked. I don't know why, I'm just saying what I did.
The way npm delivers a package is everything at once (in a tgz) - there is no possibility of it grabbing the wrong config node. However, what I have witnessed is when another fork (named alphabetically before node-red-contrib-mssql-plus) is that the plus version picks up the config node of the non plus version. Which is why I suggested you might have the old version installed.
Then it is likely you didn't restart node-red or the browser wasn't fully refreshed