💬 fjahr commented on pull request "ThreadSanitizer: Fix #29767":
(https://github.com/bitcoin/bitcoin/pull/29776#issuecomment-2035727894)
Code review ACK bbe82c116e72ca0638751e063bf564cd1fe5c4d5

This looks like the correct fix to me. CI failure is unrelated.

💬 achow101 commented on pull request "wallet: Construct ScriptPubKeyMans with all data rather than loaded progressively":
(https://github.com/bitcoin/bitcoin/pull/28333#issuecomment-2035728966)
Added a commit so that the fuzzer will skip the type of inputs that was causing the previous failure. This should be safe as that failure is not reachable in normal usage. It can only be reached if the user's wallet is corrupted, and in that case, the runtime error would propagate up to them so there would not be a crash.

✅ GopherJ closed an issue: "Always exit silently"
(https://github.com/bitcoin/bitcoin/issues/29783)

💬 GopherJ commented on issue "Always exit silently":
(https://github.com/bitcoin/bitcoin/issues/29783#issuecomment-2035808444)
closing as it's now more related to system instead of bitcoin. thanks for your help!

💬 tdb3 commented on pull request "test: Run framework unit tests in parallel":
(https://github.com/bitcoin/bitcoin/pull/29771#discussion_r1550751325)
Thank you for reviewing. **Would you be ok with a simplified approach of writing all unit test output to stdout, with this output being shown when unit tests fail?** This seems a bit cleaner and more consistent than selectively placing all unit test output to stdout or stderr depending on test pass/fail.


### More info
`TextTestRunner` lets us print unit test output to one stream. Previously (before this PR), `TextTestRunner` wrote all output to stderr regardless of pass/fail but doing t
...

💬 sipa commented on pull request "feefrac: avoid explicitly computing diagram; compare based on chunks":
(https://github.com/bitcoin/bitcoin/pull/29757#discussion_r1550770682)
Done.

💬 sipa commented on pull request "feefrac: avoid explicitly computing diagram; compare based on chunks":
(https://github.com/bitcoin/bitcoin/pull/29757#discussion_r1550770817)
Done.

💬 sipa commented on pull request "feefrac: avoid explicitly computing diagram; compare based on chunks":
(https://github.com/bitcoin/bitcoin/pull/29757#discussion_r1550770899)
Done.

💬 pablomartin4btc commented on issue "Node shutting down immediately":
(https://github.com/bitcoin-core/gui/issues/809#issuecomment-2036016180)
Shouldn't the title of the issue be updated? Perhaps adding " _- root cause: incorrect proxy IP address input_" or something? It'll be more useful navigating thru the issues list I think.

💬 tdb3 commented on pull request "test: Bump timeouts in feature_index_prune and wallet_importdescriptors":
(https://github.com/bitcoin/bitcoin/pull/29791#issuecomment-2036073972)
>As mentioned in the summary, running without --jobs=16 makes the tests succeed
>
Ok, this gives me a bit more confidence that this may just be a simple case of resource contention from high concurrency slowing things down. Omitting `--jobs` causes the default value of 4 jobs to be used. In general, IIRC jobs != threads of execution (since some tests run multiple nodes, etc.). I was also able to reproduce the failure of `feature_index_prune.py` with `--jobs=18` (on master).

> Your questi
...

💬 maflcko commented on pull request "test: Run framework unit tests in parallel":
(https://github.com/bitcoin/bitcoin/pull/29771#discussion_r1550988846)
> **Would you be ok with a simplified approach of writing all unit test output to stdout, with this output being shown when unit tests fail?**

Sure. That'd be fine as well

💬 maflcko commented on pull request "test: Run framework unit tests in parallel":
(https://github.com/bitcoin/bitcoin/pull/29771#discussion_r1550989523)
I was thinking about calling the script (passing) directly:

```
python3 ./test/functional/feature_framework_unit_tests.py

💬 willcl-ark commented on pull request "Don't permit port in proxy IP option":
(https://github.com/bitcoin-core/gui/pull/813#issuecomment-2036336106)
> I'd suggest to use a regex validator for the IP address, it's much simpler I think and the user won't be allow to enter any other symbols or chars (typo?) that aren't digits (plus user can't type more than 3 subnets separated by dots where currently you could and any symbols)

I'm not convinced that a large regex is "simpler" than checking if a ":" is in the string?

Thanks for the other suggestions, but I don't feel that interested in picking them here with the QML work going on, and them
...

⚠️ tajelp opened an issue: "Invalid port specified in -zmqpubrawtx when using IPC"
(https://github.com/bitcoin/bitcoin/issues/29802)
### Is there an existing issue for this?

- [X] I have searched the existing issues

### Current behaviour

This error is being returned if IPC is specified instead of tpc:

> Error: Invalid port specified in -zmqpubrawtx: 'ipc:///root/snap/bitcoin-core/common/.bitcoin/bitcoind.tx.raw

### Expected behaviour

IPC should correctly work

### Steps to reproduce

```
# cat ~/snap/bitcoin-core/common/.bitcoin/bitcoin.conf

server=1
rpcuser=<censored>
rpcpassword=<censored>
zmqpubhashtx=tcp://
...

👍 instagibbs approved a pull request: "feefrac: avoid explicitly computing diagram; compare based on chunks"
(https://github.com/bitcoin/bitcoin/pull/29757#pullrequestreview-1979056335)
ACK 43e2f6d160b75e88b39777428c2c1892b962f394

💬 instagibbs commented on pull request "AcceptMultipleTransactions: Fix workspace not being set as client_maxfeerate failure":
(https://github.com/bitcoin/bitcoin/pull/29735#discussion_r1551177145)
done

💬 instagibbs commented on pull request "AcceptMultipleTransactions: Fix workspace not being set as client_maxfeerate failure":
(https://github.com/bitcoin/bitcoin/pull/29735#discussion_r1551177864)
oops yeah, done

💬 fanquake commented on pull request "[WIP] ci: test secp256k1 MSAN asm annotations":
(https://github.com/bitcoin/bitcoin/pull/29742#issuecomment-2036588259)
> Ready for bump now that https://github.com/bitcoin-core/secp256k1/pull/1512 is merged.

Converted to a proper subtree bump.
Rewrote the PR description.

🤔 glozow reviewed a pull request: "AcceptMultipleTransactions: Fix workspace not being set as client_maxfeerate failure"
(https://github.com/bitcoin/bitcoin/pull/29735#pullrequestreview-1979232550)
ACK 14c86ba721e1a208c88ada133ba9e90e24724ea4

Verified the fuzz test catches this problem, seems to run fine with the changes. Thanks for accepting the suggestions.

📝 fanquake opened a pull request: "Update libsecp256k1 subtree to latest master"
(https://github.com/bitcoin/bitcoin/pull/29803)
Updates the libsecp256k1 subtree to https://github.com/bitcoin-core/secp256k1/commit/4b77fec67a80af41a538b6195317229c77101f87.

Part of #29742. See that PR for more details, the particularly relevant changes are:
* https://github.com/bitcoin-core/secp256k1/pull/1496
* https://github.com/bitcoin-core/secp256k1/pull/1512