Bitcoin Core Github
44 subscribers
121K links
Download Telegram
💬 Ayms commented on issue "Allow several OP_RETURN in one tx and no limited size":
(https://github.com/bitcoin/bitcoin/issues/27043#issuecomment-1460704159)
Cool, I am a bit like @ChristopherA probably , even if still operating a bitcoin node that I did compile myself (+ some slight modifications for testing), I did not work on it since years
💬 pinheadmz commented on issue "Many sendcmpct messages are sent during UpdateActiveChain()":
(https://github.com/bitcoin/bitcoin/issues/21903#issuecomment-1460734624)
> IsIBD() should be changed to return true while the Tip is still far behind the best known header

`IsInitialBlockDownload()` also checks if the timestamp in the tip is less than 24 hours old (`DEFAULT_MAX_TIP_AGE`) which I think means that at most we're probably only sending around 144 `sendcmpct` before we really need to.

I used `invalidateblock`/`reconsiderblock` to re-enter IBD and re-download the last 1000 blocks with net logging, and the only `sendcmpct` messages I saw go out were to
...
💬 pinheadmz commented on issue "Doc: Consolidate documentation of functional test parameters.":
(https://github.com/bitcoin/bitcoin/issues/17380#issuecomment-1460742502)
> Hi @jachiang, I have a few test runner improvement PRs written last July that I was thinking of opening at some point

@jonatack did this happen? ;-)
💬 ntherrien commented on issue ""error reading from database. shutting down"":
(https://github.com/bitcoin/bitcoin/issues/22426#issuecomment-1460751900)
Hi, thanks for looking into this over-a-year-old issue.

About the hardware defects suggestion, I remember trying it on two different computers and had the same issue. Likelihood of a hardware issue is very low in my opinion because im still using both computers a year later with other intensive tasks.

As for whether or not this would still be an issue in a newer version, I could not say because I abandoned that project because I could not get it running properly. I would have to try again
...
💬 pinheadmz commented on issue "Mac OS latest Bitcoin core latest release will not run as Mac OS, "you can't open the application 'Bitcoin core'"":
(https://github.com/bitcoin/bitcoin/issues/25834#issuecomment-1460754246)
I think this is essentially a duplicate of https://github.com/bitcoin/bitcoin/issues/15774
💬 jonatack commented on issue "Doc: Consolidate documentation of functional test parameters.":
(https://github.com/bitcoin/bitcoin/issues/17380#issuecomment-1460765343)
> @jonatack did this happen? ;-)

No, they are still in my local git. I don't propose most of the things I write to not have dozens of open unmerged PRs. It's hard to assess which ones to open, though :)
💬 jessebarton commented on pull request "doc: FreeBSD DataDirectoryGroupReadable Setting":
(https://github.com/bitcoin/bitcoin/pull/26741#issuecomment-1460812766)
The Tor project has this set (Default: 0) freebsd takes the same Default. Makes sense to me to follow what the Tor docs show rather than each individual project.
💬 ryanofsky commented on pull request "refactor: Remove CAddressBookData::destdata":
(https://github.com/bitcoin/bitcoin/pull/27224#discussion_r1129950744)
re: https://github.com/bitcoin/bitcoin/pull/27224#discussion_r1129250725

Thanks, fixed now
pablomartin4btc closed a pull request: "doc: update broken str util reference links on developer-notes"
(https://github.com/bitcoin/bitcoin/pull/27220)
📝 pablomartin4btc reopened a pull request: "doc: update broken str util reference links on developer-notes"
(https://github.com/bitcoin/bitcoin/pull/27220)
References to `utilstrencodings` and `lint-locale-dependence.sh` where incorrect, updating them accordingly.

Also, adding another reference to util function [`LocaleIndependentAtoi`](https://github.com/bitcoin/bitcoin/blob/master/src/util/strencodings.h#L108-L118), which is related with the updated section of the guide:

```
// LocaleIndependentAtoi is provided for backwards compatibility reasons.
//
// New code should use ToIntegral or the ParseInt* functions
// which provide parse err
...
pablomartin4btc closed a pull request: "doc: update broken str util reference links on developer-notes"
(https://github.com/bitcoin/bitcoin/pull/27220)
💬 fanquake commented on pull request "test: Use self.wait_until over wait_until_helper":
(https://github.com/bitcoin/bitcoin/pull/27226#issuecomment-1460862661)
I'm no-longer seeing #27208 with this change: `p2p_ibd_stalling.py | ✓ Passed | 951 s`.
📝 pablomartin4btc reopened a pull request: "doc: update broken str util reference links on developer-notes"
(https://github.com/bitcoin/bitcoin/pull/27220)
References to `utilstrencodings` and `lint-locale-dependence.sh` where incorrect, updating them accordingly.

Also, adding another reference to util function [`LocaleIndependentAtoi`](https://github.com/bitcoin/bitcoin/blob/master/src/util/strencodings.h#L108-L118), which is related with the updated section of the guide:

```
// LocaleIndependentAtoi is provided for backwards compatibility reasons.
//
// New code should use ToIntegral or the ParseInt* functions
// which provide parse err
...
⚠️ fanquake opened an issue: "Issue with `wallet_importdescriptors.py --descriptors` under valgrind"
(https://github.com/bitcoin/bitcoin/issues/27229)
This was running #27226 rebased on master (at the time 8d12127a9c19cb218d661a88ab9b6871c9d853b9).
```bash
256/256 - wallet_importdescriptors.py --descriptors failed, Duration: 3065 s

stdout:
2023-03-08T19:51:27.647000Z TestFramework (INFO): PRNG seed is: 266766692547859291
2023-03-08T19:51:27.648000Z TestFramework (INFO): Initializing test directory /home/ubuntu/ci_scratch/ci/scratch/test_runner/test_runner_₿_🏃_20230308_165259/wallet_importdescriptors_86
2023-03-08T19:51:44.422000Z Te
...
💬 pablomartin4btc commented on pull request "doc: update broken str util reference links on developer-notes":
(https://github.com/bitcoin/bitcoin/pull/27220#discussion_r1130023648)
ok
💬 pablomartin4btc commented on pull request "doc: update broken str util reference links on developer-notes":
(https://github.com/bitcoin/bitcoin/pull/27220#discussion_r1130024468)
updated it.
💬 MarcoFalke commented on issue "Issue with `wallet_importdescriptors.py --descriptors` under valgrind":
(https://github.com/bitcoin/bitcoin/issues/27229#issuecomment-1460877892)
Looks like node0 exits the http thread, but forgets to actually shut down?

```
node0 2023-03-08T20:02:32.703470Z [http] [httpserver.cpp:307] [ThreadHTTP] [http] Exited http event loop
node0 2023-03-08T20:06:42.411033Z [scheduler] [net.cpp:1532] [DumpAddresses] [net] Flushed 0 addresses to peers.dat 14ms
node0 2023-03-08T20:21:42.423331Z [scheduler] [net.cpp:1532] [DumpAddresses] [net] Flushed 0 addresses to peers.dat 10ms
node0 2023-03-08T20:36:42.434867Z [scheduler] [net.cpp:153
...
💬 MarcoFalke commented on issue ""error reading from database. shutting down"":
(https://github.com/bitcoin/bitcoin/issues/22426#issuecomment-1460887481)
We get a few data corruption reports regularly, however, no developer could reproduce them yet. So fixing them is hard and for debugging one can only throw blind guesses.

My next guess would be to check if Anti-Virus Software was involved.

If you can reliably reproduce this, it can help us if you shared the steps. Though, your Windows installation might largely differ from a fresh install of Windows. So if you can reproduce on a fresh install of Windows, that'd be ideal for finding the bug
...
💬 mzumsande commented on pull request "gui: use the stored CSubNet entry when unbanning":
(https://github.com/bitcoin-core/gui/pull/717#discussion_r1130039145)
Not a major issue, but this looks like it's designed to supports unbanning multiple nodes with one click, but the GUI doesn't allow me to select multiple entries when I try (unlike in the peertable where banning multiple nodes with once click works).
💬 mzumsande commented on pull request "gui: use the stored CSubNet entry when unbanning":
(https://github.com/bitcoin-core/gui/pull/717#discussion_r1130020981)
If this comment is fixed anyway (it was copied from `peertablemodel.h`), could also replace "getpeerinfo" with "listbanned".
💬 ryanofsky commented on pull request "refactor / kernel: Move non-gArgs chainparams functionality to kernel":
(https://github.com/bitcoin/bitcoin/pull/26177#discussion_r1130015697)
In commit "Decouple RegTestChainParams from ArgsManager" (66b17894185e0cda3ea7cbec77cc771a7ba41d56)

Probably better to merge the option instead of overwriting it:

```c++
if (auto value = args.GetBoolArg("-fastprune")) options.fastprune = *value;
```

This would avoid needing to hardcode the default `false` value here, and would be more more consistent with `version_bits_parameters` and `activation_heights` handling below which keep preexisting values if arguments are unset.

Alternat
...