Cosmos-sdk: Sending tokens to the same address should not be allowed

[kwunyeung]

Summary of Bug

I can send token from and to the same address.

Version

v0.38.3

Steps to Reproduce

desmoscli tx send desmos1fc3mdf0ue2f4suyg5vjj75jtaer0cl0dgqvy6u desmos1fc3mdf0ue2f4suyg5vjj75jtaer0cl0dgqvy6u 1udaric --from kwun --fee 20udaric

The result can be found here.

I expect this should be rejected before broadcasting the tx. The ValidateBasic is not checking if the from and to addresses are the same.
https://github.com/cosmos/cosmos-sdk/blob/218ec99508417636ccd47477bc7aad96ac0ac804/x/bank/types/msgs.go#L22

This will lead to unnecessary processing.

---

For Admin Use

• [ ] Not duplicate issue
• [ ] Appropriate labels applied
• [ ] Appropriate contributors tagged
• [ ] Contributor assigned/self-assigned

[anilCSE]

I wonder if this is a design choice to have no limitations from SDK, but it would be good to have atleast a flag enable/disable self transfers.

[fedekunze]

ahh, this is an old issue. I recall we didn't do the validation because it is up to them if they want to spend fees testing a transaction. The issue now is, what if the proposer accepts 0 fees? That could lead to DoS.

I'd propose adding an ante handler decorator to see if the transaction has positive fees.

[alexanderbez]

Self vs No-Self sends don't matter WRT to DoS -- I can simply create two addresses that I own and send the same funds back-n-forth for free (or even some multiple set of addresses).

I would recommend that we do not impose this limitation in the protocol, but rather have a warning at some level in the client UX.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.