The bug
The /locate
command sometimes returns wrong or offset coordinates for mineshafts.
Steps to reproduce
Create a Default world with the following seed
-2310970283287424513
Switch to Spectator mode to be able to look for mineshafts more easily
Use the
/locate
command to find a mineshaft/locate Mineshaft
→ It should print "Located Mineshaft at -472 (y?) 1032"
Use the
/tp
command to get to the returned position/tp -472 35 1032
Search for the mineshaft
→ There is no mineshaft nearby
Related issues
relates to
Attachments
Comments


Is this an older world or was it created in this snapshot?

Created in 16w41a

I also created a different one in 16w40a and tested that one. Similar problem.

Just created another new world in 16w41a. Tried looking for mineshaft using /locate Mineshaft. I end up in the Ocean with nothing on the ocean floor suggesting a Mineshaft

*Any news? And can somebody confirm this bug*

@unknown @unknown Can you please confirm?
Please attach world seed.
Can confirm.

You can get some parts of the command to work on some worlds while others don't , there is at least 1 or 2 broken on each 16w40a or 16w41a world. On this new world I can get Temple to work as it takes me to a witch's hut which is as intended. But Stronghold doesn't work.
It seems to offset too far.

Relates to MC-100170

Still broken.

This issue relates to MC-108011

I updated my Java to the latest version

Affects Version(s) 16w39a 16w39b 16w39c
But is slightly better on some worlds

@unknown, these versions are outdated and can't be added to the list
So, you don't need to check them =)

I know just informing that it has been broken since the start of the /locate command

This is CDES4 , but I have been locked out. Even resetting password doesn't work.
I can confirm 16w43a but now you don't get the cave as shown in Screenshot 1.
You dig to the Void.

Can somebody change the reporter to CDES5. As I have no access to my account as CDES4. And do the changes here. https://bugs.mojang.com/browse/MC-108801?focusedCommentId=338195&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-338195

@FVbico Great!

@FVbico Thanks Again!!!!!!!!!!!!!!!

Relates to MC-109311

Any update on a fix?

@unknown Please stop asking for fixes/assingees, it is only annoying.

YAY!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Can somebody confirm fix>?
Change needed:
Fixed Version 16w44a
I can confirm that this bug still seems to happen for mineshafts in 16w44a. There wasn't even any mineshaft nearby.

Abandoned Mineshafts are usually underground, try digging down?

Why do you think @unknown didn't? She's not a helper for doing nothing.
Dude, I know mineshafts are underground.
Seed: 6190859537362398884
Location when using /locate command: -250 75 50
Location the /locate command gave me: -520 (y?) 280
Switch to spectator mode and dive into the ground. There's no mineshaft there.

1. It's @unknown not @unknown @unknown
2. Just checking!!!!!

I always confuse those two! 😛

Can confirm:
Seed: 6190859537362398884
Location when using /locate command: -250 75 50
Location the /locate command gave me: -520 (y?) 280
Is true adding in panel to suggest problems.

In 16w44a:
Fortresses no longer spawn?
Can anyone confirm?
Nether fortress?

@unknown
Yes
Sorry!! My bad!

Broken on 1.11-pre1 again now.

Working seed:
Seed: -4192783173190542638
Broke seed:
Seed: -7830978626407556439

Affects 1.11

Text repair

Can confirm for Mineshafts, but is this a problem with other structures as well? If not please update the summary and description accordingly. Other structures were offset by maybe a maximum of 10 blocks, but I would not count that as bug.
And could you please clean up the description a little bit as well. It would be better if you provided for each bugged structure type one seed and tp coordinates to reproduce instead of screenshots.

Doing checks for all structures problems now.

@Marcono1234 Having problems finding errors with other structures except Mineshafts (the other structures had problems a while ago in 1.11 dev cycle). So I will upload a file called LocateRecord.txt (this is for a record of /locate MC-108801 errors) please edit this file by downloading and reuploading it.
This file should be made by typing this:
@Structurename:
FROM username
Seed: seed
Location when using /locate command: [x] [y] [z]
Location the /locate command gave me: [x] 99 [z]

For the moment a fix should be for Mineshafts as they can be easily found to have a problem.

Just adding adding 1.11.2
P.S (Mods!) Future Version - 1.11.2+ when it should be Future Version - 1.11.3+

I've got this too. It gave me co-ordinates very far away, passing several known mines, and when I got to the given co-ordinates, there was no mine in that location. I use a map program, so I can see mines in the area I've mapped, and there are plenty closer than this location, and none here.
For this world, I used "Hypothetical" as the seed, "locate Mineshaft" produced the co-ordinates "-648 (y?) 2056". I am at those co-ordinates, and there is no Mineshaft here. From the location I started at, the nearest Mineshaft is at approximately -67 👍 81.

Can confirm 17w06a

Can confirm 17w15a.

Fixed 1.12.2

Did you check a random seed, or the seed in the description? If you did a random one, please check the one in the description.

Found Mineshaft in the seed in the description but not at /locate (Image)
Some seeds have it correct (For example. 7737370236485453144)
UPDATE
All worlds have a Mineshaft spawn near where /locate says but /locate doesn't give the right coordinates. Better than pre-1.12.

This is highly reproducible in 1.12.2
In over ten attempts in 3 different worlds, the mineshaft is never at the location indicated by locate. Drilling to bedrock finds no trace of the mineshaft, nor does destroying 15x15x15 blocks down to bedrock (eg no wooden planks wreckage).
In these same worlds using chunkbase mineshaft-finder tool correctly located mineshafts in 5/5 attempts, immediately just by drilling a 1x1 vertical shaft at the coordinates given by mineshaft-finder.
There is an odd anomaly which may be a clue. The (false) coordinates returned by /locate have a repeating pattern if you run /locate in different parts of the world. For example, /locate might return -1016, Y, 250 one time, then 250, Y, -1016 the next time then -250, Y, 1016, then 1016, Y, 878, then -878, Y, 433.
To me that looks like some kind of memory or variable handling glitch, an off by one array index, an extra pop/push, something like that?

Less reproducible in 1.13[than 1.12]
But still not 100%
Sometimes now off by 1 block to exactly what it says
And since it now gives ~ which is your current y coor it means that sometimes you are underground. You might aswell give them the Y coor for the surface of that point.
And some in a world might work some might not - so seems to be some sort of registry error may be???

Using seed in the description and coor given by /locate.
Once teleported there I die (in creative mode) due to being out of the world.
[media]
Relates to MC-131462
Linked to wrong one @Kumasasa

Can somebody fix the incorrect link
Thanks

Cannot reproduce in 20w07a. All seeds in the description and comments work correctly.

The more general version of this bug would be MC-138887, which is still reproducible as of 2 days ago. The same bug exists in Bedrock, documented at MCPE-45526, in which /locate fails to locate the nearest stronghold or monument.