I've recently encountered two (possibly related?) bugs regarding fuel crossfeed. The first bug I encountered was on a vanilla Steam install of KSP 1.4 on MacOSX 10.13.3 (High Sierra).
In constructing an interplanetary vessel in a 150k orbit of Kerbin, I discovered that two engine pods (a fuel tank with docking ports and engines attached) were behaving differently in regards to fuel crossfeed. I sent up a pair of identical engine pods, pasted from the same sub-assembly, and docked them with the core of the ship. I found I was unable to transfer fuel from the central tank to the engine pods. The problem persisted after undocking and redocking the pair of pods. However, after undocking the pair from each other and attaching them individually, the outer pod (the pod not attached directly to the ship in the first docking attempt) was able to receive fuel from the central tank. Due to my having difficulties with multi-port docking (read: I'm not great at this game) I have not attempted further testing. I can confirm that the problem persists regardless of whether one port or both is docked. My current theory is that which side the pods are attached on determines fuel crossfeed capability despite the pods being symmetrical.
https://imgur.com/a/FHmA2
I'm now using a vanilla Steam install of KSP 1.4.1.
The second bug I encountered has to do with a fuel line ceasing to function completely. I have lifter stages of various sizes saved as sub-assemblies, and when I used an old 1.3 light lifter stage on a newer rocket, I had to add some sepratrons to keep it from exploding on staging. However, after I added the sepratrons, the fuel lines from that stage ceased to work entirely.
(Notice the 1.3 tanks and fuel lines connecting outer stages to inner stages. Image was taken after updating to 1.4.1)
What should be happening in the above screenshot is that the inner booster (unpinned window) should be full, while the outer booster (pinned window) was drained by the fuel line connecting the two stages. Instead, the inner booster is draining into the central booster as expected (which is why it has slightly less fuel than the outer booster) while the outer booster is isolated from the fuel reserves of the rest of the rocket.
This issue appeared only after attaching sepratrons to the outer stage of the lifter. It persists despite removing and replacing the fuel lines, and regardless of where the fuel lines are placed (from orange tank to orange tank, or from engine to engine).
