MD

I see they have added a new bug (MCPE-144208) specifically for mobs disappearing "on logout or when leave area". I guess the "on logout" could be some specific new thing, but "on leave area" is also an event that precedes, in some cases, our own favorite bug here #21416. I love you guys, we're like family now. But anyhow my concern, which with $2 can buy you a cup of coffee, is that that other bug gets investigated and solved and closed and us poor slobs over here continue to languish. Even though our mobs disappear "on leave area" all the time.

And really, "on logout" isn't unheard of here on 21416 either. It's maybe just not the entire mob collection that reliably goes >poof<. I guess if they can fix 14428 at least the game will go back to being as annoying as it was before 17.0. If I were a glass-half-full person, I'd say the 17.0 problems are probably just a misguided attempt to fix the overall despawn issue. They really are trying to fix it! <unconvincing smile>

@nefaeria, with all due respect some of us have been experiencing this symptom of entities disappearing for over four years. It is unfortunate that it seems to have gotten much worse with the latest update, but it is not some new thing even though there may be a new (additional?) cause. There is no practical difference between "disappearing" and "despawning". From the player's perspective, an entity that was in the world and which should remain in the world (due to interactions, naming, and so on) is nevertheless removed from the world. Whether it's due to chunks, distance, Nether portals, saves/crashes, or whatever other cause, the effect is the same (and IMO should be tracked here under this bug report).

The reports of entities being found underground are interesting, but until Mojang provides us some feedback (ha) about whether that is a specific issue unrelated to other entity disappearances, it's all just "where my mobs at, yo?" This is the bug for that.

I would add that it's to the benefit of people experiencing this issue for the first time to join all reports into this ticket, and VOTE for it. Multiple low-vote tickets are less likely to get attention than - and I'm just throwing out an unreasonably hyperbolic example here - a 600+ comment, 1000+ vote, 4-year old bug. As if that would ever happen, ha.

With all the focus on chunk borders, I wanted to point out especially to new commenters that among the hundreds of reports on this problem several mention sitting cats or wolves being lost. Crossing a chunk border during the save process is certainly one reasonable explanation for disappearing entities, but it can't be the only reason. As well, any cause which is not related to chunk borders could explain every other occurrence whereas "crossing chunk borders" can only explain some of them. Frankly after watching this issue evolve over 4+ years I think the issue is a more generic chunk saving problem, not specifically related to "entity crosses a border at the wrong time". I'm not convinced the borders really matter so much (not that me being convinced matters at all). The good ❓  news is that it seems to be happening so frequently now with 1.17.30 that it ought to be fairly easily repeatable by developers. Fingers crossed. All new commenters please vote for this issue, thanks.

@Jim Gersetich, they deleted my reply to you as well. They do not allow discussion of the bugfix here, only reports of entities despawning unexpectedly. With only one or two more years worth of nearly identical reports they'll have enough information to attempt a fix. It is likely that your post and this one will be deleted again.

My response to you prior was, in short: we should not assume that the solution is trivial, especially considering the behaviors people are witnessing. Even one of your own reported behaviors points not at the vanilla despawn algorithm, but at more complex problems.

@astrokitty, your calmness and optimism are admirable. I wish I was more like you in situations like this. Please continue with the positive vibes.

I speak from some experience: there are no bugs which are four years worth of "hard to fix". Not in waterfall, not in agile, not in DevOps. If it takes four years to find and fix a bug, then the bug isn't the problem. The problem is a failure of process, governance, and/or prioritization. This community has provided almost four years of tests, screenshots, examples, ideas, and workarounds, but we have not a single developer comment on bug-finding progress, nor a single patch directly for this bug. GoldenHelmet actually provided a sample world and repeatable steps! But...nothing. If I were more like you, I would not presume that they are more concerned with new features and "dungeons" and spin-off properties than they are about killing this game-breaking bug. Sadly I'm not more like you, so that is indeed what I presume. It's not credible to me that developers have been actively working on this bug for four years with no progress.

But I'm going to try to adopt some of your spirit, astrokitty, for the next year of waiting.

@astrokitty, it is the same bug. I lost numerous villagers and horses in single-player and multi-player (locally hosted, not realms) on Xbox One. Eventually stopped playing in 2019 or so because so much work was wasted. Game-killing bug now 4 years old.

Please be sure to vote for the bug. It doesn't seem to have much effect so far on getting it solved, but it's the only way to show the impact of the bug.

That is excellent information, thanks.

@thcrafter06, what were the steps you used for your testing? Based on the three years of comments it seems this bug is related more to despawning rules and the mob cap getting hit and then never recovering. IMO a comprehensive set of tests would need to be run to determine for certain if the bug had been fixed. But if you did that sort of test, it would be helpful to understand your process.

@Shannon - what we are referring to is the initial reporting of this bug, from April 17, 2017. There are ~160 comments stretching back that far and 360+ other related reported bugs. As Dr.Awesome said in his warning to me, it is a complex problem and has been with us for some time. I did not experience the bug when it was first reported, but was hit several times in 2018-2019 before I put the game on the shelf.

@Dr.Awesome - I take your point. The Helpers here have done a lot of different tests and posted their results. Even yours truly has posited some constructive ideas (or so I thought at the time). However I should not just post complaints. If you happen to know of any particular corner of this issue that the devs would like us to play with, I'll go play with it and try to be constructive again. This plus the #1 voted bug has kept me out of the game for a long time, but I'm fine going in and running particular tests. Feedback about what the bug might be and what it is not would be helpful, so we don't rehash the same topics (i.e. Nether portals).

Issue is almost exactly 3 years old now. Definitely a top priority for Mojang. Right after they overhaul cave bats.

Two things about the chunk border idea (which I think is a strong possibility) which make me go "hmmm..."

1) It's hard to see how multiple entities would vanish at once if it was a border issue. If the idea is that Chunk A is saved, entity X in Chunk B wanders into Chunk A and then some time later Chunk B is saved (poof X has disappeared), this would mean that people who notice 'several' villagers disappearing at once would have had all of those villagers wander over a chunk border at the same moment?

Alternatively, there must be a long (not milliseconds, more like seconds) delay between adjacent chunks being saved. Maybe the player moves just far enough away to cause Chunk A to save, but Chunk B stays active for a while until the player moves just far enough for Chunk B to then save?

I can easily see a chunk-border-related chunk save timing issue poofing entities one at a time over time. But groups of them all at once? Harder to explain.

2) Useful test: build two animal pens, one entirely in a chunk, one spanning a chunk boundry. Fill both with livestock. Run around like a madman, unloading and loading your animal pens as you move away/come back. Do this 1,000 times and report back the % of each pen that disappeared. Good luck we're all counting on you.

In addition to providing updates in this bug report, it is worthwhile to rate the game in the Microsoft Store and ensure your rating and comments reflect the negative effect this bug (and others) has on your enjoyment of the game. Due to this bug and the lack of hostile spawns (MCPE-21856) I haven't played Minecraft in probably a year. I keep up with the Jira updates, but so far it seems that the bugs aren't hurting sales so they're not really getting significant attention.

This is the most-voted-for issue, with over double the next-highest, and it is over two years old. Not likely they are ever going to fix it. Note that two 1.13 Beta versions still show as being affected by this bug. Bedrock is for players not interested in significant challenge from mobs, or in building working farms. And I say that as a former 100% console player. Gotta play Java from now on...

Still happening on version 1.14.3

Cape visibility is on.

This issue is not resolved.

For what it is worth, I had the City texture pack as a global resource default. Removing it from that list seemed to make the problem go away. The world opens up without re-downloading the pack now. The game seems to have a slightly new behavior in 1.6.1 where it will get to the main "Play" menu and then shortly switch to "Loading Resources" for a moment and then back to the main menu. No problems loading worlds after that.

This is a duplicate of https://bugs.mojang.com/browse/MCPE-37377, which I see you have also commented on. Be sure to vote for that other issue as well. Mods will, eventually, mark this one as a duplicate.

Hi Guy. Are you always internet-connected and logged into Xbox Live when you play? I'm not sure it matters, but there should be cloud saves of your worlds, if you are logged in when you play. I think there's a "Sync Old Worlds" or something like that on the UI (not at my Xbox now, sorry). Even though these aren't "old" worlds, I wonder if they live on in the cloud.

Any power interruptions or maybe internet interruptions while playing? A few bug reports here mention lost worlds due to power problems.

I play on Xbox One and <knock on wood> have yet to lose a world since 1.4.3. So I'm grasping at straws here. And now that I've said that I expect my world will be gone when I get home.

That was my understanding as well. FlnDutch has posted a .mcworld file in that thread which he says contains a working, stacked pair of villages (both producing golems at the expected rate). My Win10 machine is not easily available for playing MC so I haven't been able to try his world. I have not been able to duplicate the results on my own. I get behavior that would be expected if only the lowest village was active - only it produces golems. Villagers in the upper village will breed, however, up to the point where the villager count (total across both villages) matches the # of doors in both villages. Seems consistent with @Annie Ellico's comment.

Edit: he now reports that the upper village's population did transfer to the bottom village, as expected. He has some interesting results using water to phase the upper village in and out of existence, but it's not really a solution to this bug. So...nevermind I guess?

Also WAI during rain. At least so sez Wiki: "At night or during rain, villagers will run inside, closing doors behind them"

For what it's worth, someone (FlnDutch on the forums) has found that this is fixed in 1.6. I haven't been able to duplicate the results yet, but probably I am doing something wrong.

The relevant thread is here: https://www.minecraftforum.net/forums/minecraft/discussion/2923856-why-does-this-golem-farm-not-spawn-golems?page=4