💬 fanquake commented on pull request "fix: contrib: allow multi-sig binary verification":
(https://github.com/bitcoin/bitcoin/pull/23020#discussion_r1117057952)
```suggestion
--min-good-sigs 10
```

💬 fanquake commented on pull request "fix: contrib: allow multi-sig binary verification":
(https://github.com/bitcoin/bitcoin/pull/23020#discussion_r1117058463)
Where is the `--no-trust-builder-keys` option?

💬 instagibbs commented on issue "interpreter: split PrecomputedTransactionData::Init() or rename `force`":
(https://github.com/bitcoin/bitcoin/issues/27152#issuecomment-1443758072)
I obviously agree with myself, will let others chime in

💬 vasild commented on pull request "Handle CJDNS from LookupSubNet()":
(https://github.com/bitcoin/bitcoin/pull/27071#issuecomment-1443837012)
`8991ed2c6e...c2a86ba667`: rebase due to conflicts and address suggestion

@mzumsande, actually there are two problems - I am talking about Tor subnets and you are talking about a global variable used by the GUI. I will try to mimic the proxy in order to address the latter.

💬 fanquake commented on pull request "Wallet: estimate the size of signed inputs using descriptors":
(https://github.com/bitcoin/bitcoin/pull/26567#discussion_r1117183645)
> What do you think?

@achow101 followup thoughts here?

💬 fanquake commented on pull request "Wallet: don't underestimate the fees when spending a Taproot output":
(https://github.com/bitcoin/bitcoin/pull/26573#issuecomment-1443850979)
Moving this to draft for now, given it's based on #26567.

📝 fanquake converted_to_draft a pull request: "Wallet: don't underestimate the fees when spending a Taproot output"
(https://github.com/bitcoin/bitcoin/pull/26573)
Based on #26567. Alternative to #23502.

Currently, when estimating the size an input spending a Taproot output sill have once signed, we always assume the key path will be used. Even if there are script paths. This can lead to pretty severe fee underestimation if the script path turns out to be used in the end. So instead assume the most expensive between all script paths and the key path will be used.

This is still not ideal, as there may be a huge gap between the size of a script path sp
...

💬 vasild commented on pull request "Handle CJDNS from LookupSubNet()":
(https://github.com/bitcoin/bitcoin/pull/27071#discussion_r1117187548)
I tested this and there is one copy in either case. I guess the compiler is smart enough to avoid unnecessary copies.

📝 instagibbs opened a pull request: "Expand scantxoutset help text to cover tr() and miniscript"
(https://github.com/bitcoin/bitcoin/pull/27155)
null

⚠️ alex-lt-kong opened an issue: "Bitcoind crashed, how to debug"
(https://github.com/bitcoin/bitcoin/issues/27156)
OS: Debian GNU/Linux 11 (bullseye)
Available RAM: ~3GB

I run `bitcoind` on the machine by `~/bin/bitcoin-22.0/bin/bitcoind -daemon` as `root` (as this is a VM dedicated to `bitcoind`, no other ordinary users are added).
I make local RPC calls relatively frequently (up to 10k calls per sec) but only sequentially (i.e., the 2nd RPC call is only made if the first RPC call returned).

It usually works fine for a few days/weeks, then the process just disappears. The log is like the following (
...

💬 MarcoFalke commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443870264)
Did you check for OOM?

💬 fanquake commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443870993)
Going to assume this is related to lack-of-ram/a memory leak. You could run with `-debug` to produce more logs. You could also upgrade to 24.0.1, which may have fixed whichever bug you might be seeing.

💬 jonatack commented on pull request "Remove Sock::Get() and Sock::Sock()":
(https://github.com/bitcoin/bitcoin/pull/26312#issuecomment-1443941348)
re-ACK e536b3a67b390c3c46dce99cb7466862d3fd4863 per `git range-diff be2e748 c40c257 e536b3a`

💬 alex-lt-kong commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443945435)
> Going to assume this is related to lack-of-ram/a memory leak. You could run with `-debug` to produce more logs. You could also upgrade to 24.0.1, which may have fixed whichever bug you might be seeing.

I always download Bitcoin Core from https://bitcoin.org/en/download, is this considered the "official" site? On this site the latest version is 22.0 so I am still using 22.0.

💬 alex-lt-kong commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443949225)
> Did you check for OOM?

I thought about this, but consider that the [requirements page](https://bitcoin.org/en/bitcoin-core/features/requirements) shows that 1GB is enough, I did not investigate this possibility thoroughly. Btw, is `bitcoin.org` considered the "official" site anyway?

💬 fanquake commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443951699)
> I always download Bitcoin Core from https://bitcoin.org/en/download, is this considered the "official" site?

The website for the Bitcoin Core project is https://bitcoincore.org/en/download/. bitcoin.org currently lists an outdated version of Bitcoin Core for download.

💬 alex-lt-kong commented on issue "Bitcoind crashed, how to debug":
(https://github.com/bitcoin/bitcoin/issues/27156#issuecomment-1443959034)
>



> > I always download Bitcoin Core from https://bitcoin.org/en/download, is this considered the "official" site?
>
> The website for the Bitcoin Core project is https://bitcoincore.org/en/download/. bitcoin.org currently lists an outdated version of Bitcoin Core for download.

Okay then. Let me close the issue, upgrade to 24.0 and increase available RAM. See if the problem could be solved.

✅ alex-lt-kong closed an issue: "Bitcoind crashed, how to debug"
(https://github.com/bitcoin/bitcoin/issues/27156)

💬 jonatack commented on pull request "Handle CJDNS from LookupSubNet()":
(https://github.com/bitcoin/bitcoin/pull/27071#issuecomment-1443962823)
> I will try to mimic the proxy in order to address the latter.

In case you plan to do that in a follow-up:

re-ACK c2a86ba6674008996552830877d66a1bf888fad1 per `git range-diff be2e748 8991ed2c6e c2a86ba667`

💬 fanquake commented on issue "contrib/install_db4.sh script fails on OpenBSD 7.2 (RPi 3) (error: Unable to find a mutex implementation)":
(https://github.com/bitcoin/bitcoin/issues/26860#issuecomment-1443978351)
@m-kubik were you compiling the master branch? We have since removed this script, in favour of using depends. You can see the updated OpenBSD build instructions here: https://github.com/bitcoin/bitcoin/blob/master/doc/build-openbsd.md.