💬 fanquake commented on pull request "doc: recommend brew install python@3.7 (minimum version)":
(https://github.com/bitcoin/bitcoin/pull/27130#issuecomment-1437365005)
NACK - `python` and `python@3` are the same thing (both aliases for python 3.11, which is the current latest version). We can leave these docs as-is, and/or worry about changing them when Python 4 is (never) released. Thanks.

✅ fanquake closed a pull request: "doc: recommend brew install python@3.7 (minimum version)"
(https://github.com/bitcoin/bitcoin/pull/27130)

💬 fanquake commented on pull request "github: Switch to yaml issue templates":
(https://github.com/bitcoin/bitcoin/pull/27025#issuecomment-1437374469)
> I suppose one other question is whether having more/mandatory questions will put people off submitting issues,

If people can't be bothered to supply what we think is the minimum required information, then we will likely be unable to debug anything anyways. However, sometimes (fixable/informative) issues do come in the form of a randonly pasted debug log or similar. So if there is still an option for someone to dump something, I think this is ok.

@achow101 @glozow @MarcoFalke @hebasto any
...

💬 jonatack commented on pull request "docs: add ramdisk guide for running tests on OSX":
(https://github.com/bitcoin/bitcoin/pull/27124#discussion_r1111119346)
Unsure, but I thought "macOS" was the naming nowadays.
```suggestion
**macOS**
```

💬 jonatack commented on pull request "docs: add ramdisk guide for running tests on OSX":
(https://github.com/bitcoin/bitcoin/pull/27124#discussion_r1111119554)
```suggestion
(`8388608 / 2048 = 4096` or 4 GiB). To run the tests using the RAM disk:
```

💬 jonatack commented on pull request "docs: add ramdisk guide for running tests on OSX":
(https://github.com/bitcoin/bitcoin/pull/27124#discussion_r1111119386)
Not sure if it's needed to restate the OS here and line 119 above, plus "GiB", and no strong opinion about "ramdisk" but it looks like the most common spelling is "RAM disk" (https://en.wikipedia.org/wiki/RAM_disk).

```suggestion
To create a 4 GiB RAM disk named "ramdisk" at `/Volumes/ramdisk/`:
```

💬 jonatack commented on pull request "docs: add ramdisk guide for running tests on OSX":
(https://github.com/bitcoin/bitcoin/pull/27124#discussion_r1111119591)
```suggestion
To create a 4 GiB RAM disk at `/mnt/tmp/`:
```

💬 RandyMcMillan commented on pull request "doc: recommend brew install python@3.7 (minimum version)":
(https://github.com/bitcoin/bitcoin/pull/27130#issuecomment-1437396074)
To maintain consistency across macOS x86 and Arm64 maybe the minimum python3 version should be python@3.8



ref: https://github.com/bitcoin/bitcoin/pull/26226

💬 RandyMcMillan commented on pull request "Bump minimum python version to 3.7":
(https://github.com/bitcoin/bitcoin/pull/26226#issuecomment-1437397482)
https://github.com/bitcoin/bitcoin/pull/27130#issuecomment-1437396074

> To maintain consistency across macOS x86 and Arm64 maybe the minimum python3 version should be python@3.8
>
> 
>
> ref: #26226

💬 GaloisField2718 commented on issue "error: timeout on transient error: Could not connect to the server 127.0.0.1:8333 (error code 1 - "EOF reached")":
(https://github.com/bitcoin/bitcoin/issues/27121#issuecomment-1437414024)
Hello,

Now it's solved but initially it was only :
launch `bitcoind`
In another window launching `bitcoin-cli getblockchaininfo` command.

What weird was the absence of message : `bitcoin-cli` command not found.

Because now I think it's because I didn't call the path where is bin/bitcoin-cli.

Or maybe because the chain didn't finished to load and so I should wait.

I don't know exactly but now it's solve I will close it.

Thank you very much for your answer !!🙌

✅ GaloisField2718 closed an issue: "error: timeout on transient error: Could not connect to the server 127.0.0.1:8333 (error code 1 - "EOF reached")"
(https://github.com/bitcoin/bitcoin/issues/27121)

👍 furszy approved a pull request: "wallet: be able to specify a wallet name and passphrase to migratewallet"
(https://github.com/bitcoin/bitcoin/pull/26595)
rebase diff ACK 3e521b2e

💬 davidgumberg commented on pull request "rpc: Add a parameter to sendrawtransaction which sets a maximum value for unspendable outputs.":
(https://github.com/bitcoin/bitcoin/pull/25943#issuecomment-1437496703)
Force-pushed to address @glozow review.

💬 willcl-ark commented on pull request "github: Switch to yaml issue templates":
(https://github.com/bitcoin/bitcoin/pull/27025#issuecomment-1437499508)
> So if there is still an option for someone to dump something, I think this is ok.

It would also be possible to add a new `BLANK` issue template so that the blank issue template gets a big green button like the others (rather than needing to click the small hyperlink at the bottom), but IMO this would slightly defeat the point of requiring more info.

💬 willcl-ark commented on pull request "github: Switch to yaml issue templates":
(https://github.com/bitcoin/bitcoin/pull/27025#discussion_r1112292062)
Updated in 5527f7490

👍 john-moffett approved a pull request: "wallet: be able to specify a wallet name and passphrase to migratewallet"
(https://github.com/bitcoin/bitcoin/pull/26595)
ACK 3e521b2eb13799e11866562cfda4dc0bd3f7d6b9

💬 jonatack commented on pull request "blockstorage: do not flush block to disk if it is already there":
(https://github.com/bitcoin/bitcoin/pull/27039#discussion_r1112320245)
The new test fails on master as expected, because the block file is read-only and can't be written to.

It may be reassuring to have coverage for the other side of this change, to be sure we're still flushing when expected. If this new line is removed and no flushing is done, all the functional and unit tests still pass (I didn't run the extended functional tests).

```diff
@@ -618,7 +618,6 @@ bool BlockManager::FindBlockPos(FlatFilePos& pos, unsigned int nAddSize, unsigne
if ((int)n
...

💬 jonatack commented on pull request "blockstorage: do not flush block to disk if it is already there":
(https://github.com/bitcoin/bitcoin/pull/27039#discussion_r1112320479)
nit, fix clang-format if you touch this line
```suggestion
const bool position_known{dbp != nullptr};
```

💬 jonatack commented on pull request "blockstorage: do not flush block to disk if it is already there":
(https://github.com/bitcoin/bitcoin/pull/27039#discussion_r1112331348)
Adding some logging shows that this `generate` call is the slowest part of the added test. Questions:

- Why 300 blocks? On my laptop the test passes reliably with 239 blocks and fails reliably with 238. Fewer blocks = faster test runs.

- Can you explain how the number of blocks used was arrived at, and document that calculation along with the other magic numbers?

<details><summary>Ideas, feel free to pick and choose (maybe some of the logging should be debug instead of info)</summary>
...

💬 brunoerg commented on pull request "p2p: return `CSubNet` in `LookupSubNet`":
(https://github.com/bitcoin/bitcoin/pull/26078#discussion_r1112352046)
Makes sense! Gonna address it!