Add test coverage for bump fee calculation failure in feebumper #34902

issue rkrux opened this issue on March 23, 2026

rkrux commented at 1:33 PM on March 23, 2026: contributor
Motivation

There is no test coverage for the error here: https://github.com/bitcoin/bitcoin/blob/1a1f584360ce053f9f832e310d063d686f86c6d7/src/wallet/feebumper.cpp#L84-L86

Lack of coverage can be seen for line 85 (currently) here: https://maflcko.github.io/b-c-cov/total.coverage/src/wallet/feebumper.cpp.gcov.html

Possible solution

A functional test using the bumpfee RPC can be a possible solution. There is a wallet_bumpfee test file that has a good amunt of tests, which can be referred to come up with a solution.

Useful Skills
- Compiling Bitcoin Core from source
- Running the C++ unit tests and the Python functional tests
- Wallet transaction creation
- Handling unconfirmed UTXOs while creating and bumping wallet transactions
- Fee rate and mempool fee environment theoretical knowledge
Guidance for new contributors

Want to work on this issue?

For guidance on contributing, please read CONTRIBUTING.md before opening your pull request.
rkrux added the label good first issue on Mar 23, 2026
aya119119 commented at 2:52 PM on March 23, 2026: none

eloo I'de take this onee
furszy commented at 2:56 PM on March 23, 2026: member

@aya119119 awesome, welcome to Core. Feel free to tag me when you create the PR for it.
bittoby commented at 11:32 PM on March 25, 2026: none

@rkrux @furszy Can I also contribute, too?
MkDev11 commented at 9:09 AM on March 26, 2026: none

@aya119119 @furszy I opened a PR to fix this issue. please review it and share your feedback. 🙏
carloantinarella commented at 4:20 PM on March 26, 2026: contributor

@rkrux you are suggesting to develop a functional test to be added to wallet_bumpfee.py. I don’t think it is possible to have a cluster with more than 500 transactions without exceeding the hardcoded limit MAX_CLUSTER_COUNT_LIMIT{64} set in txgraph.h, am I right?

sendrawtransaction RPC should return too-large-cluster.
cprkrn referenced this in commit e991d5be3e on Mar 31, 2026
cprkrn referenced this in commit 81348f5a38 on Mar 31, 2026
cprkrn referenced this in commit b5ab51102d on Mar 31, 2026
cprkrn referenced this in commit c6198f4bda on Mar 31, 2026
cprkrn referenced this in commit b5dbe4abdc on Apr 1, 2026
reza6ix commented at 3:29 AM on April 4, 2026: none

I'd like to take a look at this. I'm currently setting up my local build with bitcoind and the test framework. I have experience with [C++/Python] and I've been contributing to other high-scale projects like Dify.

I'll investigate the current test coverage in feebumper.cpp and see where the failure case needs to be added to the functional tests. I'll follow up with a PR or questions once I have the reproduction running
LucasVarella commented at 8:53 AM on April 7, 2026: none

Hi, @rkrux & @furszy! If you could review my PR I would be extremely grateful ;D
maflcko removed the label good first issue on Apr 7, 2026
maflcko commented at 9:13 AM on April 7, 2026: member

Removing the good first issue label, because it encourages llm bots to create duplicates