Interesting position evaluation - Why is White evaluated as +1 or more

I had this interesting position in the Caro-Kann. What makes it interesting for me is that black is +2 pawns, has the better pawn structure, has a passed pawn and yet the computer is evaluating it as +1.36 for white.
If I play it out using a computer at max on both sides, it ends in a draw.

I am curious as of why the computer evaluates this as so bad for black, I think there is something to learn for me here.

EDIT: to clarify, I understand it, as "The positional advantage of white is so big, that it makes up 3!! pawns."
I want to understand what the positional advantage is, that white has.
I see that white has castled, and a bit more active pieces, but black also has an intact castling side and the king did not move yet.

I did check with the computer. I also saw that he at some point forces a passed pawn. But i don't see anything suspicious in this position, that warrants a 3 pawn compensation. You say massive attack and too many threats. Please enqlighten me which ones. It's black turn. I only see two haging pawns that seemingly are fine after b6.

Well, this IS a horrible position for Black. The king's rook and knight are temporarily stuck, the king is badly placed, and the queenside is vulnerable. Saying that white has "a bit more active pieces" is ridiculous, since you are way behind in development and even your Be7 is taking Ng8's best square for development.

The mandatory move to keep the game alive is 17...Kf8 (the only way to avoid instant collapse after Rxb7), but still Black has huge problems after 18.Rxb7 or probably 18.Rc1.

What do you like about this horrible mess of a position? Even if a computer somehow finds an acrobatic way to survive, a real person is very unlikely to find a long sequence of "only moves" to avoid disaster.

Interesting. I mean , sure i'm underdeveloped being busy eating enemy pawns. However I did not expect this underdevelopment to be more worth than that.

I really try to grow here, to learn to see what you are seeing.

I understand now the predicament of the stuck knight and rook, which also blocks the castling. I did not realize that. So now I wonder, theoretically, if the knight could develop somehow, more quickly allow castling, would this already massively shift things?

The necessity is not losing your whole Queenside, and there aren't many ways to do that- only 17...Kf8 which unpins the knight (and creates the problem of a rook at h8 which will be inactive for a long time).
17...b6?? loses a piece after 18. Rc1 (or Rc3).

17..Rc7 still loses the b7 pawn after 18.Rfb1, although 18.Rxb7 Rxb7 19.Bxc6+ Rd7 20.Rb1 may be even stronger.

Just out of curiosity, I fired an engine to check 17...Kf8 at considerable depth, and it came out with something "reasonable" at +0.8 after 18.Rc1 Rd8! (only move) 19.Bxc6!? d4! (only move again) and so on.

But you do need to find some extraordinary defensive moves to achieve this, so you may well dismiss the whole line as too risky and impractical.

Hmm, thanks for the input!

The interesting thing is the question how to know when to skip gabbing a seemingly free pawn because it would slow development. Not obvious for me yet. I mean I fight hard to create such opportunities in the first place. And then I shoukd skip them because I'm not yet fully developed? I'll try to look out for this.

And again, as the noob I am I would have chosen to play as black here without doubt

+2 in material
passed pawn
better pawn stucture (enemy has an isolated edge pawn)
castling rights still intact

I now see that my rook and knight are basically stunned for some time and my castling rights only exists on paper. Thanks for pointing that out. But hard for me to see that this is such an disadvantage, just by judging the current position. But your input helped to clear this a bit. Thanks pfren & Josh

Hkjgihvc