Identical Parsley

This is now in 26.3. As an update, my nether portal links correctly if I enter from a specific direction moving to a specific block. Entering from the other side, or from the other two possible blocks causes it to generate a new portal in the aforementioned location approximately 50x 130z blocks away from the intended portal.

On the nether side, the portal is 2 wide, 3 tall facing north/south. Going from -46,97,-127 to -46,97,-128 causes the portal to link back correctly. Going from -46,97,-129 to -46,97,-128 causes it to incorrectly generate a new portal. Going from -47,97,-127 to -47,97,-128 causes it to incorrectly generate a new portal. Going from -47,97,-129 to -47,97,-128 causes it to incorrectly generate a new portal.

On the overworld side, the portal is 3 wide by 4 tall, facing north/south, centered on -298,78,-891. The incorrectly generated new portals all face north/south as expected, and are being generated gradually further in the -Z direction as I take them down but leave the trenches they create in the grass.

My seed is -5131095867450498466.

At an incorrectly generated portal, looking at the trenches of the others and toward where the expected destination portal is.

The starting portal on the overworld side, and the expected destination when returning.

The initially autogenerated portal on the nether side, looking at the one block which correctly links back as expected.

Higher view of how the incorrectly generated overworld return portals were placed, gradually being placed further and further north.

In 26.3, I’m getting behavior similar to this, except my portals are not within 128 blocks of zero in either direction. I initially built my portal at -298, 78, -891. This sent me to Nether -45, 97, -128. Returning through that portal sends me to approximately -366, 69, -1026. Going back in returns me to the same nether location. Coming back out returns me to the same unwanted Overworld location, away from the portal I built but within sight of it. If I take down the portal and repeat the process, another new unwanted Overworld portal is created in approximately the same location plus or minus a few x/z blocks.

Exactly once in my testing, the portal worked as expected going from Nether to Overworld brought me to my correct initially built portal. This bug makes a great way to get lots of obsidian though!

Same. The crash still occurs when I try to load 26.2-snaphsot-2. My device cannot run Vulkan either.

Now in 1.21.11 the behavior is still present. I thought to give the pillagers the slow falling effect so the fall would not be fatal to them. As expected they land onto the pressure plates and activate them correctly. Seems like this is more evidence that the problem is that they are dying and despawning before the pressure plate checks to see if something hit it.

I recreated the world using the same seed, transplanted my -3 -1 region file and then walked to the tower. That worked. The modified tower structure is present and the backup is only ~60MB instead of ~600. This is now 1.21.5. The problem is still present. I observed exactly one instance of a pillager setting off a pressure plate at the bottom of the pit with a number of others falling in and not setting them of.

Do you also need the level.dat file?

Sorry, how do I do that? I can give you the world backup, but it’s 600MB. I don’t know how to extract just the tower structure. I can give you the .mca file the tower is located in. That’s only 9MB. Is that sufficient?

Finally, here is an occasion when it worked as expected. The pillager fell in, the pressure plate activated, and the piston caused the note block to play.

Here is the view with the debugging information turned on.

Here is the view from the top with a pillager having just hit the bottom and died.

Yes, nothing has been done to affect their spawn rate. They spawn, come to the tower any typically fall into the trench one at a time, occasionally two or three at a time. They drop only the ordinary and expected loot.

I have not experimented with other kinds of mobs falling into the trench.

Here is a view from the top looking down into the trench with a pillager in the process of falling in. When he hits the pressure plates at the bottom he visibly dies and despawns but the pressure plate usually does not activate. They activate perhaps 5% of the time when pillagers fall on them. When I jump into the trench, they always activate, and I presume this is because the height is not fatal to me.