Testing Bitcoin Core 23.0 Release Candidates (tests)

Mar 30, 2022

Host: stickies-v

Notes

Major versions of Bitcoin Core are released every 6-8 months. See the Life Cycle documentation for full details.
When all of the PRs for a release have been merged, Release Candidate 1 (rc1) is tagged. The rc is then tested. If any issues are found, fixes are merged into the branch and a new rc is tagged. This continues until no major issues are found in an rc, and that rc is then considered to be the final release version.
To ensure that users don’t experience issues with the new software, it’s essential that the rcs are thoroughly tested. This special review club meeting is for people who want to help with that vital review process.
This Bitcoin Core Release Candidate Testing Guide has a tutorial for testing the release candidate.
- It is recommended to go through the “Preparation” steps ahead of the meeting, especially if you want to compile from source.
- The testing guide relies on the tools jq and watch, which are not installed by default on each platform. For example on macOS, you can install these ahead of time using brew install. Alternatively, you can also modify the instructions to avoid using these tools as they are not strictly necessary and/or can be replaced by other tools.
- For one of the tests, we try to connect over CJDNS. If you have not used this before, you could save some time by going through the instructions and configuring this ahead of the meeting.
The guide is just to get you started on testing, so feel free to read the Release Notes and bring ideas of other things you’d like to test!

Meeting Log

13:00

<stickies-v> #startmeeting

13:00

<glozow> hi’

13:00

<jaonoctus> hello!

13:00

<brunoerg> hi

13:00

<ccdle12> hi

13:00

<svav> Hi

13:00

<hernanmarino_> hello !

13:00

<stickies-v> welcome to a special edition of the review club everyone!

13:00

<justin_> hey :)

13:00

<monlovesmango> hey

13:00

<danielabrozzoni> hi 🙂

13:00

<noc_> hi

13:01

<larryruane> hi

13:01

<kouloumos> hi

13:01

<lightlike> hi

13:01

<GucciPoet> Hi guys, I am honored to be here for my first meeting. You guys are all Legends!!! I'm here to learn and grow as a developer. nice to meet everyone!

13:01

<josibake> hi

13:01

<brunoerg> GucciPoet: welcome!

13:01

<stickies-v> welcome GucciPoet, we're glad to have you hear! do we have any other people joining for the first time?

13:01

<jaonoctus> GucciPoet: welcome!

13:01

<stickies-v> *here, ouch

13:01

<emzy> hi

13:02

<stickies-v> with the upcoming release of bitcoin core v23.0, today we'll be going through the testing guide that tests some of the changes introduced in this version

13:02

<stickies-v> you can find the meeting notes (incl link to the guide) here: https://bitcoincore.reviews/v23-rc-testing

13:03

<stickies-v> the guide is meant as a *starting point* to get you going and cover some of the important changes. it's crucial that enough people test different paths of the software, so after the hour is over - go nuts and try to break things!

13:03

<stickies-v> (and please report your findings)

13:03

<stickies-v> before we start, could you please share which OS you're using and whether you'll be compiling from source or using the precompiled binaries?

13:03

<stickies-v> macOS 12.3 (arm64) / compiled from source

13:04

<danielabrozzoni> I'm on NixOS 21.11, compiled from source

13:04

<brunoerg> macOS 12.0 / compiled from source

13:04

<emzy> macOS 12.3 / compiled from source

13:04

<ccdle12> nix + ubuntu / compiled from source

13:04

<monlovesmango> macOS 11.2 (arm64) / precompiled binaries

13:04

<jaonoctus> Ubuntu on Windows hehe, 5.10.16.3-microsoft-standard-WSL2 / compiled from source

13:04

<kouloumos> macOS 10.15.7 / compiled from source

13:04

<josibake> pop_os 21 (ubuntu) / compiled from source

13:05

<hernanmarino_> Linux 64 bits / compiled from source

13:05

<GucciPoet> macOS 10.15.6/ compiled from source

13:05

<stickies-v> jaonoctus great to see some windows users too, please don't hold back on running the windows build if you have time!

13:05

<larryruane> Linux 64 Ubuntu 21.10, compiling from source

13:06

<jaonoctus> stickies-v: surething, will do l8r

13:06

<stickies-v> looks like most people are quite comfortable with compiling from source, but feel free to use the binaries too if you run into any compilation problems

13:07

<stickies-v> so let's get started and get our environment setup, as per https://github.com/bitcoin-core/bitcoin-devwiki/wiki/23.0-Release-Candidate-Testing-Guide#preparation

13:07

<ls55> Fedora 35 / compiled from source

13:07

<stickies-v> we'll give a few minutes for people to get their release candidate, but no need to wait on everyone else if you'd like to start working through the sections or have questions

13:07

<larryruane> don't have to rabbit-hole but i'm having trouble compiling from source (i haven't seen this before): `/usr/bin/ld: /g/db4/lib/libdb_cxx-4.8.a(os_yield.o): in function '__os_yield':`

13:07

<larryruane> `os_yield.c:(.text+0x79): undefined reference to pthread_yield`

13:08

<larryruane> i think it may be because i'm on ubuntu 21.10, which may be too new to be supported(?)

13:08

<josibake> larryruane: looks related to bdb

13:09

<josibake> larryruane: try setting the --without-bdb flag in configure?

13:09

<larryruane> ok will do, thanks (i was compiling with BDB)

13:09

<glozow> larryruane: maybe just try a autogen, configure, make again? 😅

13:10

<stickies-v> the build process should become a bit smoother for everyone as we're moving away from bdb, which is nice!

13:10

<stickies-v> alright if you run into any issues with compiling or running the binaries do shout for help, but let's move on to the first test

13:11

<stickies-v> just to make sure everything boots alright, we're doing a quick wallet test in the GUI: https://github.com/bitcoin-core/bitcoin-devwiki/wiki/23.0-Release-Candidate-Testing-Guide#testing-the-gui

13:11

<brunoerg> great

13:13

<jaonoctus> gui is working here :)

13:13

<josibake> anyone else successfully build cjdns from source? im getting an error

13:14

<ls55> Yes, I did. Can you describe the error ?

13:14

<larryruane> glozow: josibake: (building without dbd worked, thanks)

13:14

<monlovesmango> gui works for me. small question, why can't you load wallet backup files from gui? can that be done from command line? (sorry im new)

13:14

<stickies-v> TIL Qt is pronounced "cute" instead of "cutie". What a day!

13:15

<emzy> I did only a basic GUI test like sync, receive and send. Worked.

13:15

<jaonoctus> stickies-v: fr? TIL+1

13:15

<stickies-v> that's a great question monlovesmango, and no need to apologize! does anyone have an answer?

13:15

<larryruane> GUI starts up for me too, looks good!

13:15

<brunoerg> GUI working here!

13:16

<stickies-v> jaonoctus: https://en.wikipedia.org/wiki/Qt_(software)

13:16

<hernanmarino_> GUI working. Syncing , sending and receiving working too

13:16

<monlovesmango> for cjdns I ended up using homebrew (brew install sjdns) and it seemed to be running, but also couldn't get cjdns IBD to work

13:16

<glozow> monlovesmango: see https://github.com/bitcoin-core/gui/pull/471

13:16

<monlovesmango> glozow: thanks!!

13:16

<larryruane> would someone like to post a signet address here? i'll send to it using "cute" (haha, qt)

13:17

<ls55> There is a PR that allows users to restore GUI wallet backup from GUI

13:17

<GucciPoet> Everything loaded, created wallet..works

13:17

<danielabrozzoni> larryruane (IRC): tb1qa6shvf2p8nvfz30a9ykepq8l8ue9258kntu9kv 🙂

13:18

<josibake> no issues with the gui, nice that descriptors are default now :)

13:18

<emzy> tb1p8w0fx4h63mmxxhc6az8s7qugtqzrvyrtmjtz9agx958e4x5x5j9ql0xgme

13:18

<stickies-v> can't say no to free coin, yes please larryruane tb1q8cjpapvzyayxkhqnmjcfx8stshtl5gtdv3czu5

13:18

<monlovesmango> I will try to test that pr after this :)

13:18

<hernanmarino_> emzy: i 'll send you some sats

13:18

<emzy> ^ taproot address

13:18

<danielabrozzoni> Sats received, thanks!

13:19

<larryruane> danielabrozzoni: sent (0.4)

13:19

<stickies-v> monlovesmango re CJDNS did you try the ping6 command from the guide to check you're connected?

13:19

<emzy> hernanmarino_: thx. I see the TX

13:20

<hernanmarino_> emzy : great

13:20

<stickies-v> alright looks like the GUI is working well for everyone, time to move on to a slightly tricker test

13:20

<monlovesmango> stickies-v: yes and couldn't get ping6 to work for anything but localhost. but i realized that my router wasn't setup for ip6 so waiting for bf to stop working to try again

13:20

<danielabrozzoni> I had troubles with cjdns as well... cjdroute started without errors, but pinging wouldn't work

13:21

<larryruane> emzy: sent to you (0.1 to ...xgme)

13:21

<emzy> WOW I got a second TX with 0.1

13:21

— emzy feels signet rich ;)

13:21

<stickies-v> so far most nodes have been running on the default port of 8333, but there is an ongoing effort to make it more feasible to run on most available ports, including #23542 which removes Core's very strong preference to only connect to peers running on the default port

13:22

<stickies-v> since there's randomness involved the test in https://github.com/bitcoin-core/bitcoin-devwiki/wiki/23.0-Release-Candidate-Testing-Guide#testing-non-default-ports can take a very short or a very long time, so don't panic if it takes a while!

13:23

<monlovesmango> this one passed for me. but the monitoring script had some weird output in the 'Currently connected outbound peers:' report section

13:23

<stickies-v> what was the weird output?

13:24

<jaonoctus1> thanx @amitiuttarwar for that PR #22618

13:24

<larryruane> stickies-v: sent 0.15 to ...czu5 (all my sends have been using gui)

100

13:24

<emzy> I also got a nonstandard port peer. Was after about 15 minutes.

101

13:24

<danielabrozzoni> I had a weird output as well but stickies-v fixed it ~6 hours ago 😀 See https://github.com/bitcoin/bitcoin/issues/24706#issuecomment-1082971761

102

13:25

<monlovesmango> id column had: /tmp/23-rc-test/connection_monitoring.sh: line 3: /bitcoin-cli: No such file or directory

103

13:26

<stickies-v> it looks like you didn't set your $BINARY_PATH properly then, because it's pointing to your $DATA_DIR instead

104

13:26

<emzy> monlovesmango: yes you have to set export BINARY_PATH=$(pwd)/src also

105

13:26

<stickies-v> well it's not necessarily pointing to your DATA_DIR but I suppose it was pointing to somewhere wrong

106

13:26

<monlovesmango> danielabrozzoni: ok cool ill try again

107

13:27

<monlovesmango> i was using precompiled binary in downloads folder, so BINARY_PATH=/Users/monica/Downloads/bitcoin-23.0rc2/bin/

108

13:27

<emzy> I assume BINARY_PATH is just not set. Had the same problem.

109

13:27

<stickies-v> can anyone explain why we couldn't speed up this test by having everyone here run their node on a non-default port and then see if we're able to connect to it?

110

13:28

<stickies-v> emzy monlovesmango yeah just run echo $BINARY_PATH and see if that points to your binary dir

111

13:28

<larryruane> in case anyone's not aware, you can send yourself some signet funds from the faucet here: https://signet.bc-2.jp/ (i just did that and the gui detected the incoming payment)

112

13:29

<josibake> stickies-v: this isn't really testing if we can/cant connect on non-default ports, rather its testing that addrman connects to non-defaults when we become aware of them, right?

113

13:29

<josibake> previously, we had a strong preference for connecting to nodes on the default port (8333)

114

13:29

<monlovesmango> stickies-v: echo $BINARY_PATH looks ok. i'll try again later and post results

115

13:30

<glozow> josibake: yep. testing automatic, not manual, connections to non-default ports

116

13:30

<stickies-v> josibake indeed! manual connections were possible already

117

13:30

<stickies-v> (automatic connections too, but extremely rare)

118

13:31

<stickies-v> from the monitoring script, i'm curious what's everyone's AddrMan non-default to default ratio?

119

13:31

<stickies-v> and if you connected to a non-default peer, what's your ratio of connection attempts made?

120

13:31

<monlovesmango> monitoring script is working now!

121

13:32

<stickies-v> my AddrMan ratio is 4.57%, and connection attempts currently at 4.82% (but not yet finished)

122

13:32

<emzy> stickies-v: I had 5%

123

13:33

<monlovesmango> AddrMan: 5.14%, connection attempts: 6.12%

124

13:34

<stickies-v> Test successful: connected to non-default peer on "139.162.22.75:18222" whooo I passed!

125

13:36

<stickies-v> "It does not necessarily mean 5% of our peers are non-default, since e.g. there could be biases like non-default nodes being less reliable than default nodes."

126

13:36

<stickies-v> Does anyone have any ideas as to what kind of biases that could be?

127

13:37

<danielabrozzoni> While running the test, I sometimes get a `error code: -29 Node not found in connected nodes`... is it normal?

128

13:38

<stickies-v> danielabrozzoni yes this happens sporadically when the peer already disconnects by itself right before we run the disconnectnode command

129

13:38

<emzy> I also got that.

130

13:38

<danielabrozzoni> Ah, makes sense, thanks 🙂

131

13:40

<stickies-v> I'm not sure if everyone found their non-default peer already, but as we don't have that much time left I'd suggest we move on to CJDNS next

132

13:40

<svav> I just ran bitcoin-23.0rc2-win64-setup.exe and the Welcome Window has really small text and is hard to read ...

133

13:40

<josibake> was able

134

13:40

<stickies-v> https://github.com/bitcoin-core/bitcoin-devwiki/wiki/23.0-Release-Candidate-Testing-Guide#testing-cjdns

135

13:40

<josibake> to get a non-default peer! kudos on the script stickies-v

136

13:41

<stickies-v> does anyone have any ideas why we'd bother adding CJDNS when we already have Tor and I2P?

137

13:41

<stickies-v> (or before that, can anyone briefly summarize what CJDNS is about?)

138

13:41

<stickies-v> yay josibake !

139

13:42

<sipa> more mechanisms for combatting partitioning helps :)

140

13:42

<stickies-v> svav that's good to know, would you be able to briefly check if there's already an issue for that and create one otherwise? https://github.com/bitcoin/bitcoin/issues

141

13:43

<monlovesmango> from the little I could understand, cjdns is like a homegrown network of vpns...?

142

13:44

<stickies-v> sipa yeah absolutely! And if you look at this from an adversarial angle, we need to make sure to have alternatives. If our privacy depends on just one network, it would be easier to bring down. With support for multiple privacy networks, we have much more resiliency

143

13:45

<GucciPoet> agreed

144

13:45

<josibake> from the little reading on cjdns , it appears it was originally designed to be used as a mesh network. not sure if this was part of the motivation for adding it, but seems like that could also be very useful for bitcoin

145

13:46

<sipa> i'm not sure privacy is really the argument here... being on more networks sounds strictly worse for privacy

146

13:46

<sipa> but privacy isn't the only concern

147

13:46

<svav> stickies-v ok I will do that

148

13:47

<svav> How do I stop the IBD on Windows, because I don't have room for the whole download on this computer?

149

13:47

<stickies-v> (note: it's always useful to look at the PR when trying to understand the reasons for implementing a change, there's often a healthy discussion. For example, see https://github.com/bitcoin/bitcoin/pull/23077 for CJDNS)

150

13:48

<stickies-v> I barely know anything on CJDNS myself so I'm also just asking the questions here haha, can't give too much answers

151

13:48

<emzy> I think more networks helps against eclipse attacks.

152

13:48

<josibake> svav: its fine to let it run while you are working through the guide. you can delete the tmp datadir when you are done

153

13:48

<svav> oh ok thanks josibake

154

13:48

<stickies-v> svav you can stop it with `$BINARY_PATH/bitcoin-cli -datadir=$DATA_DIR stop`

155

13:49

<josibake> emzy: correct, which is why its not recommended to run only tor, or only i2p, or only cjdns. very few nodes using those networks makes it easier to eclipse

156

13:49

<stickies-v> how's everyone getting along with setting up CJDNS? any issues with pinging the sample peer?

157

13:49

<stickies-v> ping6 fc32:17ea:e415:c3bf:9808:149d:b5a2:c9aa

158

13:52

<josibake> stickies-v: still running into compile errors for cjsdns, but ill finish the cjdns testing steps later once i get this figured out

159

13:52

<danielabrozzoni> I can't ping the sample peer ☹️

160

13:53

<stickies-v> josibake are there no binaries available or ubuntu?

161

13:53

<lightlike> josibake: i'm not sure if tor-only is still a problem today, since there has been such a huge increase in available peers.

162

13:53

<stickies-v> danielabrozzoni did you add a couple of peers from hyperboria-peers and then restart your cdjroute? (note: you have to run it with sudo!)

163

13:55

<danielabrozzoni> So, I added IPv4 peers only, as I run into trouble when adding IPv6 peers (but I don't have a IPv6 IP, so I guess it makes sense)

164

13:55

<danielabrozzoni> And then yeah, I restarted with sudo, but no luck

165

13:55

<danielabrozzoni> It shows no error when starting, the log looks fine

166

13:56

<josibake> lightlike: good point!