fix(bmr): send only one receipt per transaction

[MuhammedIrfan]

Closes #543
Closes #186

Checklist:

• I have performed a self-review of my own code
• I have documented my code in accordance with the documentation guidelines
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• I have run the unit tests
• I only have one commit (if not, squash them into one commit).
• I have a descriptive commit message that adheres to the commit message guidelines

[codecov]

Codecov Report

Merging #545 (bf923b0) into main (4e691bb) will increase coverage by 0.49%.
The diff coverage is 77.27%.

@@             Coverage Diff              @@
##               main     #545      +/-   ##
============================================
+ Coverage     42.01%   42.51%   +0.49%     
  Complexity      801      801              
============================================
  Files           115      116       +1     
  Lines         11321    11317       -4     
  Branches        941      941              
============================================
+ Hits           4757     4811      +54     
+ Misses         6155     6093      -62     
- Partials        409      413       +4     

Impacted Files	Coverage Δ
cmd/iconbridge/chain/icon/sender.go	0.00% <0.00%> (ø)
cmd/iconbridge/chain/bsc/sender.go	33.85% <50.00%> (-4.86%)	⬇️
cmd/iconbridge/chain/near/sender.go	61.49% <66.66%> (+14.00%)	⬆️
cmd/iconbridge/common/chainutils/segment.go	89.28% <89.28%> (ø)
cmd/iconbridge/chain/bsc/verifier.go	47.05% <0.00%> (-0.66%)	⬇️
cmd/iconbridge/chain/substrate-eth/receiver.go	62.77% <0.00%> (ø)
cmd/iconbridge/chain/bsc/receiver.go	59.02% <0.00%> (+1.42%)	⬆️

[bbist]

Hi @MuhammedIrfan,
Can you try with the latest changes? You need to provide a configurable value for tx_data_size_limit from the config file as specified here.

[CyrusVorwald]

These changes need tests.

[bbist]

These changes need tests.

Thanks @CyrusVorwald-ICON for pointing it out. It wasn't a final solution, but rather an intermediate one to see if it works for NEAR. Anyway, I'll refactor the shared code for segmentation used across all chains and add necessary test and push again.

[bbist]

Hi @MuhammedIrfan,

The tests that were failing on near side were not implemented correctly. I've changed the test parameters so that they work as expected. They should work now.

[bbist]

Hi @MuhammedIrfan,

The tests that were failing on near side were not implemented correctly. I've changed the test parameters so that they work as expected. They should work now.

Btw, have you deployed this code? Let's test it on local and see if it resolves the original issue that this PR resolves. If it does, I'll approve the changes and merge into main.

[MuhammedIrfan]

@bbist looks like it does not add message to tx if rlp size is more than configured limit also if it have only 1 event

[bbist]

@bbist looks like it does not add message to tx if rlp size is more than configured limit also if it have only 1 event

Yeah. That's true. You can configure a limit that can handle 1 event. We did the same for BSC.

[MuhammedIrfan]

@bbist looks like it does not add message to tx if rlp size is more than configured limit also if it have only 1 event

Yeah. That's true. You can configure a limit that can handle 1 event. We did the same for BSC.

But we can't always be sure of size right, at times when the event size is more than configured it create relay message without any receipts eg: https://explorer.testnet.near.org/transactions/3or4kyeL7fQ6uds73BfCynMbHscnBG4tLjwCHNA4EJon

[bbist]

@bbist looks like it does not add message to tx if rlp size is more than configured limit also if it have only 1 event

Yeah. That's true. You can configure a limit that can handle 1 event. We did the same for BSC.

But we can't always be sure of size right, at times when the event size is more than configured it create relay message without any receipts eg: https://explorer.testnet.near.org/transactions/3or4kyeL7fQ6uds73BfCynMbHscnBG4tLjwCHNA4EJon

What is the maximum number of events the NEAR chain can handle in a single transaction? Is it a single event of can it handle 2-3 or more? If so, you can chose some data size limit that can cover 2 or more events. And in the worst case, it'll forward 1 event. This is what we did for BSC.
If this approach is sufficient, then it's fine. Otherwise, we might have to modify the code to also support for minimum number of events, rather than the transaction data size limit.

[MuhammedIrfan]

@bbist

in the worst case, it'll forward 1 event. This is what we did for BSC. If this approach is sufficient, then it's fine. Otherwise, we might have to modify the code to also support for minimum number of events, rather than the transaction data size limit.

This is not happening, I have updated the test to test this scenario. Please check.

What is the maximum number of events the NEAR chain can handle in a single transaction? Is it a single event of can it handle 2-3 or more?

At present NEAR supports only one event which is being improved as part of gas optimization, As different BTP Messages have different gas usage, for example, Init BTP Message will use less than that transfer, and Link BTP Message uses more gas, Until we find a way to optimize and segment the message in the relay we might just keep only 1 BTP Message per Relay Message

[bbist]

@bbist

in the worst case, it'll forward 1 event. This is what we did for BSC. If this approach is sufficient, then it's fine. Otherwise, we might have to modify the code to also support for minimum number of events, rather than the transaction data size limit.

This is not happening, I have updated the test to test this scenario. Please check.

What is the maximum number of events the NEAR chain can handle in a single transaction? Is it a single event of can it handle 2-3 or more?

At present NEAR supports only one event which is being improved as part of gas optimization, As different BTP Messages have different gas usage, for example, Init BTP Message will use less than that transfer, and Link BTP Message uses more gas, Until we find a way to optimize and segment the message in the relay we might just keep only 1 BTP Message per Relay Message

Well, in that case as a temporary solution, let's have a minimum number of events set to 1 while segmenting for NEAR.

[bbist]

@bbist

in the worst case, it'll forward 1 event. This is what we did for BSC. If this approach is sufficient, then it's fine. Otherwise, we might have to modify the code to also support for minimum number of events, rather than the transaction data size limit.

This is not happening, I have updated the test to test this scenario. Please check.

What is the maximum number of events the NEAR chain can handle in a single transaction? Is it a single event of can it handle 2-3 or more?

At present NEAR supports only one event which is being improved as part of gas optimization, As different BTP Messages have different gas usage, for example, Init BTP Message will use less than that transfer, and Link BTP Message uses more gas, Until we find a way to optimize and segment the message in the relay we might just keep only 1 BTP Message per Relay Message

Well, in that case as a temporary solution, let's have a minimum number of events set to 1 while segmenting for NEAR.

@MuhammedIrfan,
Has necessary changes been made on this issue? Please don't let it be a blocker for the integration. You can make necessary changes in your NEAR specific Segment() function to forward a single event, and maybe add a TODO comment to comeback to this issue later. Let's try to resolve this PR.

[andrii-kl]

It would be good to generate a random one instead predefined for private key