How to find out why a document get is slow?

I have also found that strange, when I've checked the docs which said they were deprecated in 2.3.0.
Here's what I have:

curl -s 'http://localhost:9200/_cat/nodes?v&h=v,id'

v id
5.1.2 9YnN
5.1.2 wzS5
5.1.2 mTac
5.1.2 -_zl
5.1.2 uZ2l
5.1.2 Zhv2
5.1.2 ebJt
5.1.2 HEI1
5.1.2 sPrn
5.1.2 Gwgn
5.1.2 USXH
5.1.2 kxzX
5.1.2 Cksl
5.1.2 H3w1
5.1.2 XdeO
5.1.2 q8Wx
5.1.2 1qbk
5.1.2 X_QX
5.1.2 GauH
5.1.2 43Gj
5.1.2 1T6O
5.1.2 Wyje
5.1.2 b6L9
5.1.2 HC5i
5.1.2 g6XG
5.1.2 BojW

BTW, the indexes were created in 2.x.