Thanks, this time I see all the nodes in the list except the data nodes. I'm guessing you'll see the same even with multiple runs of the query.
This tells me that the UI is doing the right thing, in that it's displaying the nodes available in the monitoring data. So now we can narrow down the scope of this problem to the "backend" — why no monitoring data is being collected for data nodes.
Let's focus on one specific data node, say esd1. If we can get that to show up in the list, chances are whatever fixes we had to make for it could be applied to the remaining data nodes as well.
Would it be possible for you to look at the Elasticsearch server logs for esd1? You don't have to copy/paste them all here. Just look for any errors or warnings, especially with words like "monitoring", "collection", "collector", in them. If you find any such errors or warnings, please post them here (or pastebin if there's a lot).
Thanks!