search

pit-coin / pit-coin.github.io

PITcoin Bond
http://pitcoin.network
0 stars 2 forks source link

bug bounty #1

Open pit-coin opened 4 years ago

pit-coin commented 4 years ago

Find bugs or backdoors in the ERC20 ethereum token.

Github: https://github.com/pit-coin/pit-coin.github.io/blob/master/contracts/Token.sol

Ropsten testnet: https://ropsten.etherscan.io/address/0x6342A5c056F71E7E3a6Bf89560Dc1F97210bDb51

gitcoinbot commented 4 years ago

Issue Status: 1. Open 2. Started 3. Submitted 4. Done

This issue now has a funding of 0.1 ETH (24.64 USD @ $246.37/ETH) attached to it.

gitcoinbot commented 4 years ago

Issue Status: 1. Open 2. Started 3. Submitted 4. Done

Work has been started.

These users each claimed they can complete the work by 266 years, 3 months from now. Please review their action plans below:

1) l-kh has started work.

I can do an automatic scan for your contract. 2) nightwolf3 has started work.

I've been a blockchain dev for about two years and would love to help you debug the contract 3) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 4) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 5) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 6) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 7) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 8) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 9) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 10) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 11) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 12) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 13) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 14) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 15) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 16) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 17) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 18) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 19) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 20) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 21) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 22) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 23) abdel-az has started work.

I am enthousiaste in blockchain and i am interessed to this challenge 24) srisankethu has started work.

I have found some issues in the contract 25) ashish10677 has started work.

I'll go through the code thoroughly and then try to run it locally and check for issues. 26) surajsingla333 has started work.

I have good experience in smart contract development and debugging and will be able to do a thorough analysis. 27) cryptaldev has started work.

trying to break ERC20 contract and siphon finds 28) barryjo has started work.

recently started working on defi space and have good knowledge of ERC standards 29) cryptoicarus has started work.

Reviewing code and will attempt to exploit bug on testnet 30) jordan69420 has started work.

I'll review your code, and see if it has any technical or legal problems

Learn more on the Gitcoin Issue Details page.

L-KH commented 4 years ago

I am not that good to detect hidden bugs. and this rapport is what I get after using MythX dev plan. HIGH-SWC-101 | The arithmetic operation can underflow. line 47 & line 40

description: It is possible to cause an arithmetic overflow. Prevent the overflow by constraining inputs using the require() statement or use the OpenZeppelin SafeMath library for integer arithmetic operations. Refer to the transaction trace generated for this issue to reproduce the overflow.

Instructions to reproduce the vulnerability of line 40(Test Case 1)

Instructions to reproduce this vulnerability line 40

You can check JSON file that I export from MythX: MythXDevScan.zip

hope it was helpful to you.

srisankethu commented 4 years ago

@pit-coin Giving a glance, I found 2 issues:

  1. Shouldn't require be used instead of assert in add, mul, signedAdd and signedSub? Criteria for using require and assert here.
  2. There is no check here if b==0 in div. Check SafeMath.sol. Possible break at here.
gitcoinbot commented 4 years ago

Issue Status: 1. Open 2. Started 3. Submitted 4. Done

Work for 0.1 ETH (24.38 USD @ $243.78/ETH) has been submitted by:

  1. @l-kh
  2. @cryptoicarus

@pit-coin please take a look at the submitted work:

pit-coin commented 4 years ago

@L-KH, I suppose it should work correctly. I wrote some tests.

srisankethu commented 4 years ago

@pit-coin I made a comment on 2 issues I saw

pit-coin commented 4 years ago

@srisankethu

  1. This checks for overflow, which should not happen under any circumstances, so I use assert. I use require in sub because it checks for valid input.
  2. Solidity reverts tx if divide by zero. It is possible only if single owner transfers all tokens. In this case all the gas would be consumed.
CryptoIcarus commented 4 years ago

There are a few issues I was able to find.

  1. Recommend making the contract owned and the clean() function onlyowner, currently anyone can withdraw any token from contract

  2. In buy function use if statements instead of require. Using require will result in unnecessary errors.

function buy(address _ref) public payable {
        // w*pr = T + T*ppt/m - P + R
        // with ref
        // in*pr = t + f + r
        // (w + in)*pr = T+t + (T+t)*(ppt + f*m/T)/m - (P + t*(ppt + f*m/T)/m) + (R + r)
        // no ref
        // in*pr = t + f
        // (w + in)*pr = T+t + (T+t)*(ppt + f*m/T)/m - (P + t*(ppt + f*m/T)/m) + R
        // first
        // in*pr = t
        // (w + in)*pr = T+t + (T+t)*ppt/m - (P + t*ppt/m) + R

        uint256 tokens = msg.value.mul(price);
        uint256 fee = tokens.div(10);
        tokens = tokens.sub(fee);

        _ref = _ref != msg.sender ? _ref : address(0);
        if (_ref != address(0) && balanceOf[_ref] >= refRequirement) {
            uint256 refBonus = fee.mul(3).div(10);
            fee = fee.sub(refBonus);
            refDividendsOf[_ref] = refDividendsOf[_ref].add(refBonus);
        }

        uint256 increaseProfitPerToken = 0;
        if (totalSupply != 0) {
            increaseProfitPerToken = fee.mul(multiplicator).div(totalSupply);
            profitPerToken = profitPerToken.add(increaseProfitPerToken);
        } else {
            tokens = tokens.add(fee);
        }

        balanceOf[msg.sender] = balanceOf[msg.sender].add(tokens);
        totalSupply = totalSupply.add(tokens);
        emit Transfer(address(0), msg.sender, tokens);

        uint256 payout = tokens.mul(profitPerToken).div(multiplicator);
        payoutsOf[msg.sender] = payoutsOf[msg.sender].signedAdd(payout);

        emit Buy(msg.sender, _ref, tokens, increaseProfitPerToken);
    }
  1. Very minor bug. It is possible to sell a very small amount of tokens with 0 fees and not increasing profitPerToken. Link to tx exploiting below. 0x3fade616c50e7f3997f942818151baac7a779fe840b2ab329ef4d0be235ee549

Either add: 

require(_tokens >= 10)

or add/change:

require(_tokens > 0);
uint256 fee = _tokens >=10 ? _tokens.div(10) : 1;
pit-coin commented 4 years ago

@CryptoIcarus, I appreciate a lot what you have done!

  1. The idea is that the contract does not have an owner at all, so, yes, anyone can withdraw tokens sent to this contract by mistake, for advertising or else.
  2. This is incorrect input or an attempt to cheat, i think, so it log an error.
  3. Yes, it is possible. I suppose no one would do this because the fee is greater than expected profit. I consider this as an improvement, not as a bug.
CryptoIcarus commented 4 years ago

@CryptoIcarus, I appreciate a lot what you have done!

  1. The idea is that the contract does not have an owner at all, so, yes, anyone can withdraw tokens sent to this contract by mistake, for advertising or else.
  2. This is incorrect input or an attempt to cheat, i think, so it log an error.
  3. Yes, it is possible. I suppose no one would do this because the fee is greater than expected profit. I consider this as an improvement, not as a bug.
  1. No, not necessarily attempts at cheating. These would likely be user errors.

Example 1: The code below throws an error if the balance of the refer is too small. However it should not be the job of the user who is buying to check the balance of the refer address every time. E.g. User 1 buys tokens once with user 2's ref link and it works. Then user 2 sells tokens and no longer meets balance requirements. User 1 now tries to buy tokens again using the ref link and it fails. They would see this on their end as a bug and not know how to fix the issue.

require(balanceOf[_ref] >= refRequirement, "small balance");

Example 2: This could be an attempt at cheating but could also be a simple user error. I assume that users are sharing ref links and not manually entering refer addresses. E.g. User 1 buys tokens. User 1 creates a ref link to share with user 2. User 1 then wants to buy more tokens, however they may have bookmarked their own ref link or they have visited the site more times with the ref link and their browser autocompletes to the ref link. Again the transaction fails and the end user would not understand why.

require(_ref != msg.sender, "_ref is sender");

These are not hypothetical examples. I have personally written contracts before that include refer functions and users very often submit addresses that do not meet requirements unknowingly. Most users do not have the knowledge to go look on etherscan and check what the error was. Read the code and figure out how to remedy the situation.

gitcoinbot commented 3 years ago

Issue Status: 1. Open 2. Started 3. Submitted 4. Done

The funding of 0.1 ETH (59.44 USD @ $594.38/ETH) attached to this issue has been approved & issued to @cryptoicarus.