Testing Bitcoin Core 26.0 Release Candidates (tests)

Nov 15, 2023

Host: m3dwards

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 provides guidance 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. Verify and confirm the successful installation of v26rc2.
- For some of the tests, you might want to have the signet chain fully synced beforehand, so that you can just copy the signet directory into /tmp/26-rc-test/ every time you run a test in a fresh environment.
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

17:00

<stickies-v> #startmeeting

17:00

<stickies-v> hi everyone!

17:00

<maxedw> hi

17:01

<lightlike> hi

17:02

<stickies-v> anyone here for the first time? even if you're just lurking, feel free to say hi!

17:04

<stickies-v> not a massive turn out, it seems

17:05

<sipa> hi, lurking

17:05

<d33r_gee> hello

17:05

<stickies-v> ah, more people, yay!

17:06

<stickies-v> we've got a special edition of review club planned for this week, ahead of the (hopefully soon) final release of Bitcoin Core v26

17:07

<stickies-v> maxedw has prepared a fantastic guide to help test (some of) the new features of this new release: https://bitcoincore.reviews/v26-rc-testing

17:07

<stickies-v> the purpose of this meeting really is just to go through it together

17:08

<stickies-v> given that we're a bit of a smaller group, just polling if everyone's still keen to do that in this meeting, or if people would prefer to just go through it on their own pace?

17:08

<maxedw> thank you stickies-v

17:08

<maxedw> hello everyone, thanks for joining us for today's review club

17:08

<maxedw> we will be taking a look and a run through the Bitcoin Core 26.0 testing guide

17:08

<maxedw> link to the guide: https://github.com/bitcoin-core/bitcoin-devwiki/wiki/26.0-Release-Candidate-Testing-Guide

17:09

<maxedw> what operating systems are people running?

17:09

<d33r_gee> WSL Ubuntu 22.04

17:09

<lightlike> I will do some testing later this week at my own pace, so I'd be just lurking.

17:09

<maxedw> no problem lightlike

17:10

<maxedw> d33r_gee: great stuff

17:10

<stickies-v> i'm on macos 13.4, m1 chip

17:10

<maxedw> great stickies-v, there is a macos only test that you will be able to help out on

17:10

<maxedw> today we will cover: getprioritisedtransactions rpc, mportmempool rpc, V2 Transport - BIP 324, TapMiniscript and Ancestor Aware Funding

17:11

<maxedw> I will leave the last test, Outbound connection management for you to do in your own time after this meeting

17:11

<maxedw> Did anyone manage to run through the preparation steps? No problem if not.

17:11

<d33r_gee> yep

17:12

<maxedw> fantastic, we won't need to spend much time getting prepped then

17:13

<stickies-v> same here!

17:13

<stickies-v> running compiled binaries

17:13

<maxedw> let's start the getprioritisedtransactions rpc test together

17:13

<stickies-v> well, self compiled

17:13

<maxedw> this should be a relatively easy test to start us off

17:15

<stickies-v> test works fine for me!

17:16

<d33r_gee> test went through no problems

17:16

<stickies-v> well i did find something interesting

17:16

<stickies-v> if i run the prioritisetransaction rpc twice on the same tx, the feerates compound

17:17

<stickies-v> right, okay, that is in line with the documentation of the rpc: "The fee value (in satoshis) to add (or subtract, if negative)"

17:18

<maxedw> great to hear, it also worked on my machine

17:18

<maxedw> interesting stickies-v

17:18

<stickies-v> https://github.com/bitcoin/bitcoin/blob/0aa014d5a34ed6b020b687ec924f8a17351f5aeb/src/rpc/mining.cpp#L455

17:18

<maxedw> if everyone is ready, let's move onto the second test

17:18

<d33r_gee> ready

17:19

<maxedw> importmempool

17:19

<maxedw> for this test, we can do it on mainnet (preferred) or on regtest.

17:19

<maxedw> does anyone have a mainnet node synced that they can test this RPC with?

17:21

<stickies-v> yup

17:21

<maxedw> great

17:22

<d33r_gee> yup got one that was synced using an utxo snapshot

17:22

<maxedw> that's excellent

17:22

<maxedw> nice to see mainnet being used where possible

17:22

<maxedw> let's have a go at this one

17:24

<maxedw> you may need to pay attention to the environment variables $DATA_DIR to make sure you have it pointing to a bitcoind data directory that's synced

17:25

<d33r_gee> when I first ran importmempool got this error:

17:25

<d33r_gee> error code: -10

17:25

<d33r_gee> error message:

17:25

<d33r_gee> Can only import the mempool after the block download and sync is done.

17:25

<d33r_gee> However waited a bit then reran and it worked

17:27

<maxedw> hmm, d33r_gee. I might have expected with `connect=0` it never to sync if it thought it was out of sync

17:27

<maxedw> but I think that's still a successful test

17:29

<d33r_gee> maxedw ah great I test it some more, do you think it may be because the chainstate came from a snapshot?

17:29

<maxedw> I'm not sure tbh

17:29

<maxedw> wouldn't have thought so

17:29

<d33r_gee> maxedw np

17:30

<stickies-v> d33r_gee: what do you get when you run `bcli getblockcount`?

17:31

<maxedw> I think I might add a check into the test to make sure we are fully synced before we stop bitcoind

17:32

<d33r_gee> maxedw great idea

17:32

<stickies-v> oh wait you're saying it worked eventually

17:32

<maxedw> at the moment mine says 816730

17:33

<d33r_gee> stickies-v yet waiting like 2min then reran and it worke

17:33

<stickies-v> my importmempool is still running, seems to take quite a while for a big mempool

17:33

<maxedw> but mine is stuck from my previous test with connect=0

17:34

<maxedw> stickies-v: for me it didn't take long

17:34

<maxedw> did it take a while for you d33r_gee ?

17:35

<d33r_gee> maxedw yep took longer than the regtest one for sure

17:35

<stickies-v> how big was yours maxedw? I've got ("bytes": 1716322) and I'm on a pretty old machine

17:35

<maxedw> "bytes": 8098791

17:37

<stickies-v> interesting, looks like it's not really working for me then

17:39

<maxedw> let's pick this up after the call stickies-v

17:39

<maxedw> I just ran mine again and it completed in about 10 seconds

17:40

<stickies-v> according to `top` it seems to be doing something, running with -debug=rpc and -debug=mempool but it doesn't really give me much, there's no intermediate logging

17:40

<stickies-v> i'll just let it run and see what happens

17:40

<maxedw> great idea

17:40

<maxedw> this is exactly why we run these tests!

17:41

<maxedw> let's move on to the next test

17:41

<stickies-v> oh wait

17:41

<stickies-v> i have an interesting idea to test

17:41

<stickies-v> on mainnet

17:41

<maxedw> go ahead stickies-v

17:42

<stickies-v> what happens when you run importmempool, abort, run again, in very short sequence?

17:42

<maxedw> fyi, after my node synced my mempool size was similar to yours

17:42

<stickies-v> (abort the cli request w ctr+c, not kill the daemon)

17:42

<maxedw> great question

17:43

<maxedw> I can give it a go. d33r_gee are you still in a position to try?

17:43

<d33r_gee> maxedw yep

17:43

<stickies-v> my import finally succeeded: "Imported mempool transactions from disk: 40705 succeeded, 0 failed, 0 expired, 0 already there, 0 waiting for initial broadcast"

17:46

<d33r_gee> running importmempool right now

17:46

<maxedw> I reran the test with ctrl+c and then imported again and it worked for me

100

17:46

<maxedw> appears to anyway

101

17:46

<d33r_gee> it worked

102

17:46

<d33r_gee> "size": 48876

103

17:47

<maxedw> Imported mempool transactions from disk: 0 succeeded, 0 failed, 0 expired, 2959 already there, 0 waiting for initial broadcast

104

17:47

<maxedw> that is my second log message

105

17:47

<maxedw> so ctrl+c didn't appear to stop the import for me

106

17:47

<maxedw> I was quite quick at hitting it but maybe still too slow

107

17:47

<maxedw> or perhaps the command is running in another thread that doesn't exit

108

17:48

<stickies-v> no ctrl+c on the cli doesn't affect the daemon, and we don't recognize remote client disconnects

109

17:48

<stickies-v> so bitcoind keeps processing all requests even if the client's no longer listening

110

17:48

<stickies-v> but i was wondering if running simultaneous importmempool tasks would cause issues

111

17:48

<stickies-v> great, looks like it doesn't

112

17:49

<maxedw> I was perhaps not quick enough for it to be simultaneous

113

17:49

<maxedw> I will have another go after this meeting

114

17:49

<maxedw> shall we move onto the next test?

115

17:49

<maxedw> V2 transport

116

17:50

<d33r_gee> sounds good

117

17:50

<maxedw> which again could be done on different networks but for this guide we are using signet

118

17:51

<maxedw> I believe bitcoin.achow101.com should support v2 protocol on both signet and mainnet so please feel free to use whichever network you prefer

119

17:52

<maxedw> I just tried bitcoin.achow101.com on mainnet and got: Error: v2transport requested but not enabled (see -v2transport)

120

17:53

<maxedw> so I think if you plan to use that node, it's signet only at the moment

121

17:53

<stickies-v> mmm i think that's an error on your side maxedw

122

17:53

<maxedw> you are right

123

17:54

<maxedw> I skipped a step

124

17:54

<maxedw> (didn't enable it for myself)

125

17:58

<maxedw> I now have: New manual v2 peer connected: version: 70016, blocks=816906, peer=18

126

17:59

<stickies-v> hmmm i wasn't able to connect to achow but i think he just wasn't accepting new inbounds

127

17:59

<stickies-v> will investigate more

128

17:59

<maxedw> the peer isn't listed in my peerinfo either

129

17:59

<maxedw> but I had that message in the logs

130

17:59

<maxedw> looks like we are out of time to run through the last couple of tests

131

18:00

<maxedw> #endmeeting