This is happening for me on bevy 0.1.3. The behavior appears to be introduced by commit 4eb437. The immediately prior commit (2dadc8) produces the behavior I expect.
Calling commands.spawn is causing one of my entities to disappear, and causes another entity to take on the Transform of the entity that disappeared. I suspect that some kind of entity ID or entity index is becoming incorrect.
Here is a sample of correct behavior with my code running with bevy#2dadc8:

He is a sample of the incorrect behavior with bevy#4eb437:

Notice on the client that the cube disappears and that the plane appears to be getting the Transform that should be being applied to the cube.
Finally, on 4eb437 if I comment out spawning the messages that display on the sidebar in the client, the correct behavior is restored (except the client no longer has messages in the sidebar list)
// commands.spawn((Message::new(
// &server_message.message,
// &server_message.from,
// server_message.ordinal,
// ),));

Reproduction
This can be reliably reproduced by:
ncallaway/bevy_prototype_networking_laminar commit 90cebc (currently master)cargo.toml to point bevy at the last known working commit:bevy = { git = "https://github.com/bevyengine/bevy.git", rev="2dadc86fb003f64db5e6870a327fd0ed969881e5" }
cargo run --release --example testbed -- -s 127.0.0.1:12530, cargo run --example testbed -- -c 127.0.0.1:12541 127.0.0.1:12540 fooWASD and the cube position is replicated to the client)cargo.toml to point bevy at the next commit:bevy = { git = "https://github.com/bevyengine/bevy.git", rev="4eb437ab2e3c2b69ade01f3661fe128fe8027daa" }
examples/testbed/net/mod.rs lines 127-131Suggested Labels: bug | ecs
not sure if this is related, but I experienced a similar thing where the material would "jump" between entities even if I don't have any systems that change the material, and each entity has their own material. Not working on materials atm so I haven't had the chance to investigate it yet.

This started happening after updating to 0ae74a4a4dd3acfeebb2d6e6040c25fad70a2c9f, I was using 2dadc86fb003f64db5e6870a327fd0ed969881e5 before this.
I have a feeling that this is related to #299, which i am investigating now. Probably caused by #216
@cart Yea, that looks similar to the commit that introduced the issue for me (4eb437).
@sanisoclem Any chance you could try the two commits I identified as bad and good and see if your issue was introduced in the same commit?
Here's a minimal repro example, with some of my findings in case they help. https://github.com/jngbsn/bevy_entity_bug
@sanisoclem Any chance you could try the two commits I identified as bad and good and see if your issue was introduced in the same commit?
yep, confirming that the issue only happen when using 4eb437ab2e3c2b69ade01f3661fe128fe8027daa and up.
Cool cool. Thanks y'all! I'm working on a fix now.
This issue no longer repros for me with master after the fix was merged!
@jngbsn @sanisoclem you might double check your test cases with master, but I suspect this is fixed.
fixed for me as well, despawn works as expected too. Cheers!
Thanks for checking!
Most helpful comment
This issue no longer repros for me with master after the fix was merged!
@jngbsn @sanisoclem you might double check your test cases with
master, but I suspect this is fixed.