Uploaded image for project: 'Minecraft: Java Edition'
  1. Minecraft: Java Edition
  2. MC-147152

Invalid move player packet received / teleporting player to NaN crashes worldsave


    • Confirmed
    • Commands, Crash, Entities

      In short:
      At least ender dragons summoned without a DragonPhase seem to be able to "bug out" and be positioned at x=NaN, y=NaN, z=NaN.
      If a player teleports to that dragon, the worldsave crashes and you get the info:
      "[Server thread/INFO]: <Playername> lost connection: Invalid move player packet received"

      Upon trying to reenter, you enter the world very briefly, then get kicked with the same message, and your worldsave will be unplayable.

      No crash report gets produced.
      Video proof

      To check on my bugposts related to ender dragons, I entered one of my according bugtest worldsaves.
      In this worldsave is a (non-moving) dragon already waiting, summoned with /summon ender_dragon ~ ~10 ~

      This time I did not see it and at first thought it was a visual problem (see https://www.youtube.com/watch?v=-7UESPgjZME)

      I teleported to the dragon with /tp @s @e[type=minecraft:ender_dragon,limit=1], not knowing that it was at position NaN.
      Then the crash occured:

      [Server thread/INFO]: Meri lost connection: Invalid move player packet received

      Worldsave is unplayable now, but I always make backups upon entering with a new MC version.

      Last time I had entered this worldsave was in 19w07a.
      I don't know what changed since then that the Dragon will be now located at NaN/NaN/NaN if it was not summoned with a DragonPhase.
      This also evidently happens to not only newly summoned dragons, but also those which have been already summoned in an older worldsave.


      [15:35:13] [Server thread/INFO]: Scanning for legacy world dragon fight...
      [15:35:13] [Server thread/INFO]: Found that the dragon has been killed in this world already.
      [15:35:13] [Server thread/INFO]: Found that there's a dragon still alive (arw['Ender Dragon'/1515, l='Dragon Test', x=NaN, y=NaN, z=NaN])

      This crash/bug is reproducible with the attached worldsave Dragon Test_Crash-Reproduction.zip.

      How to reproduce

      • Enter worldsave
      • Note that there is a dragon boss bar, but you can't see a dragon.
      • Teleport to dragon:
        /tp @s @e[type=minecraft:ender_dragon,limit=1]

        You'll lose the connection/get kicked with "Invalid move player packet received"

      • Try to reenter the world, you'll be able to, but only for a very short time.
        Note that in the log it'll look similar to this:
        <Playername>  logged in with entity id 165 at (NaN, NaN, NaN)

      and you'll get kicked immediately again.

      Screenshot of player position NaN


      Side note:
      I don't know why, but when I summon an ender dragon and then kill it, it removes 9 entities. No matter if the dragon got summoned with or without a DragonPhase. (MC-146503)
      Haven't searched yet if this was already reported; display of 9 killed entities confirmed by Panda4994.
      /say @e[type=ender_dragon] also has 9 outputs, and all of them also have a different UUID, even though you only summoned 1 dragon before.
      This happens also in freshly created worlds, be it the Legacy dragon or a freshly summoned one, when you execute the kill-command, it will find several dragons with different UUIDs and you'll be able to execute the kill-command several times. The amount seems to vary, so far it was either 6 or 7 or 9 times for me.

        1. 2019-03-31_16.47.04.png
          3.36 MB
        2. 2019-03-31_16.58.06.png
          2.97 MB
        3. Dragon Test_Crash-Reproduction.zip
          4.11 MB
        4. image-2019-03-31-17-06-41-470.png
          176 kB
        5. MC147152.mp4
          7.95 MB
        6. screenshot-1.png
          21 kB
        7. screenshot-2.png
          23 kB
        8. ticking-entity.png
          97 kB

            dinnerbone [Mojang] Nathan Adams
            LapisDemon Meri Diana
            1 Vote for this issue
            2 Start watching this issue