💬 hebasto commented on issue "Unable to open wallet UI with ubuntu 23.10":
(https://github.com/bitcoin/bitcoin/issues/29311#issuecomment-1910533393)
On Ubuntu 23.10, I've installed the Bitcoin Core 26.0 from the "latest/stable" Snap channel.

> The terminal output of the command line
> `Fontconfig warning: FcPattern object weight does not accept value [0 205) [1] 824648 segmentation fault (core dumped) env /snap/bin/bitcoin-core.qt %u`

Cannot reproduce it.

> Hello Thank you this command line works [keshavbhatt/olivia#95 (comment)](https://github.com/keshavbhatt/olivia/issues/95#issuecomment-774747492)

Indeed.

> Close or move
...

💬 sdaftuar commented on issue "Cluster mempool, CPFP carveout, and V3 transaction policy":
(https://github.com/bitcoin/bitcoin/issues/29319#issuecomment-1910540024)
cc: @TheBlueMatt @rustyrussell @Roasbeef @t-bast

💬 real-or-random commented on pull request "build: Pass sanitize flags to instrument `libsecp256k1` code":
(https://github.com/bitcoin/bitcoin/pull/28875#discussion_r1466616567)
> I wasn't too worried.

Yep, I agree, any approach is ok in the end. My feeling is just that setting `SECP_CFLAGS` is not the cleanest choice and more likely than my other suggestions to cause breaks in the future.

💬 Retropex commented on pull request "Add a `-permitbarepubkey` option":
(https://github.com/bitcoin/bitcoin/pull/29309#issuecomment-1910542474)
tACK https://github.com/vostrnad/bitcoin/commit/8c1114aa61c46e58efb44368b5428d3ccb65f9d0

Steps to test:

1. Build [vostrnad branch](https://github.com/vostrnad/bitcoin/tree/permitbarepubkey)
1. Start a regtest with `-chain=regtest`
1. Create a wallet with `bitcoin-cli createwallet`
1. Get address with `bitcoin-cli getnewaddress`
1. Generate blocks to get bitcoins `generatetoaddress 500 <address obtained previously>`
1. Create a P2PK tx, this [repository](https://github.com/dianerey/bec
...

💬 Sjors commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466618909)
Mmm, I tried something similar in de26cbc79230f64d7747b8b88f4dee71ebd37e1b by adding a `DOCKER_RUN` variable and setting it to `podman run --replace`. But this fails with `Error: short-name resolution enforced but cannot prompt without a TTY`. Are you passing any other arguments in your alias?

🤔 pablomartin4btc reviewed a pull request: "debugwindow: update session ID tooltip"
(https://github.com/bitcoin-core/gui/pull/788#pullrequestreview-1844140068)
Concept ACK

nit: @MarnixCroes, maybe you can add the screenshot for transport v1 and session ID is empty/ not shown, if possible, even this PR doesn't affect that logic.

💬 fanquake commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466623922)
> Are you passing any other arguments in your alias?

The alias came from:
```bash
apt install podman-docker
# podman-docker is already the newest version (4.8.3+ds1-2).
docker --version
# podman version 4.8.3
podman --version
# podman version 4.8.3
```

💬 fanquake commented on pull request "ci: Use LLVM 18.x & DEBUG=1 in depends for MSAN jobs":
(https://github.com/bitcoin/bitcoin/pull/27495#discussion_r1466634562)
```bash
CMake Error at /msan/llvm-project/libcxxabi/CMakeLists.txt:51 (message):
LIBCXXABI_USE_LLVM_UNWINDER is set to ON, but libunwind is not specified in
LLVM_ENABLE_RUNTIMES.


-- Configuring incomplete, errors occurred!
```

💬 fanquake commented on pull request "ci: Use LLVM 18.x & DEBUG=1 in depends for MSAN jobs":
(https://github.com/bitcoin/bitcoin/pull/27495#discussion_r1466637149)
Added libunwind the the libc++ build as well

💬 Sjors commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466637332)
podman-docker is not documented in the manual installation... https://podman.io/docs/installation#building-from-source

Let's see how I can install this without totally messing up my OS :-)

💬 achow101 commented on pull request "wallet: guard against dangling to-be-reverted db transactions":
(https://github.com/bitcoin/bitcoin/pull/29253#issuecomment-1910567093)
> seems like something I would never expect to happen unless someone modified the code or manually corrupted the data on disk. So I'm wondering did we accidentally cause a breakage like that during development of these PRs?

It shouldn't ever happen.

In general, we don't have very robust error handling for when the database fails to read or write something. I think this should actually be more like an assert, but one that doesn't kill the whole software, maybe just unloads the wallet?

💬 maflcko commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466643431)
You can try:

```sh
mkdir -p ~/bin/ && echo 'IyEvdXNyL2Jpbi9lbnYgYmFzaAoKZWNob2VycigpIHsgZWNobyAiJEAiIDE+JjI7IH0KCmVjaG9lcnIgIiAoOjo6ZG9ja2VyLT4pcG9kbWFuICR7QH0iCgpwb2RtYW4gIiR7QH0iCg==' | base64 --decode > ~/bin/docker && chmod +x ~/bin/docker

💬 glozow commented on pull request "policy: enable sibling eviction for v3 transactions":
(https://github.com/bitcoin/bitcoin/pull/29306#discussion_r1466648992)
Do you mean all of the sibling eviction logic? That doesn't work because we'd only enforce that the tx pays enough to evict siblings OR conflicting transactions; it needs to pay for both.

If you mean the logic adding stuff to `m_conflicts`, `m_iters_conflicting`, etc., it's because I don't like in-out parameters.

💬 Sjors commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466651006)
That base64 decodes to:

```sh
#!/usr/bin/env bash

echoerr() { echo "$@" 1>&2; }

echoerr " (:::docker->)podman ${@}"

podman "${@}"
```

That doesn't seem any different from my `DOCKER_RUN="podman run --replace"` alias.

💬 Sjors commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466652244)
Or I guess it's the `echoerr` doing some magic

💬 instagibbs commented on pull request "policy: enable sibling eviction for v3 transactions":
(https://github.com/bitcoin/bitcoin/pull/29306#discussion_r1466657588)
the later, but at least you gave a reason why not

💬 glozow commented on pull request "policy: enable sibling eviction for v3 transactions":
(https://github.com/bitcoin/bitcoin/pull/29306#discussion_r1466658531)
I think it's impossible to hit, since we're v3-only and it can't have another ancestor. Maybe we just omit this case? I figured somebody would advocate for having it just in case.

💬 glozow commented on pull request "v3 transaction policy for anti-pinning":
(https://github.com/bitcoin/bitcoin/pull/28948#issuecomment-1910596660)
Last push incorporated a rework to `PackageV3Checks` from @sdaftuar (thanks), addressing a problem [discussed in irc yesterday](https://bitcoin-irc.chaincode.com/bitcoin-core-dev/2024-01-24)

💬 maflcko commented on pull request "Support self-hosted Cirrus workers on forks (and multi-user)":
(https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1466665081)
It is hard to debug your issue, but I presume you have docker installed on your system.

In this case, you can't install podman-docker without removing docker. Also, the CI system workers will not clean up after themselves, because the cleanup is done with the `podman` command, which can not clear docker containers.

My recommendation would be to start from a fresh VM (fresh install of Ubuntu) and then follow the docs: https://github.com/bitcoin/bitcoin/blob/ac923e70e7cec603abd207f104dbabfe6
...

👍 ryanofsky approved a pull request: "refactor: Compile unreachable walletdb code"
(https://github.com/bitcoin/bitcoin/pull/29315#pullrequestreview-1844216896)
Code review ACK fa3373d3adbace7e4665cf391363319a55a09a96. This looks good, and should prevent code in the else blocks from accidentally breaking.

Maybe a way to make it a little clearer would be to define `use_bdb` and `use_sqlite` c++ constants like:

```
constexpr bool use_bdb{
#ifdef USE_BDB
true
#endif
};
```

Then the actual code could use `if constexpr(use_bsb)` like a normal conditional. This might be more verbose than the current approach, though, and the current approach do
...