v27.0 Testing

[marcofleon]

Umbrella issue for 27.0 testing. Please help testing on a wide variety of supported platforms, as well as interaction with different software.

Let us know which version you tested on which operating system.

If you find an issue, please search Github for known issues first and then open a new Github issue.

This meta issue should not be used to report bugs, as a single thread makes it impossible to track more than one topic.

See 27.0 Release Notes Draft for a list of changes.

See here for the testing guide: https://github.com/bitcoin-core/bitcoin-devwiki/wiki/27.0-Release-Candidate-Testing-Guide

[jonasnick]

Tested 27.0rc1 on NixOS without encountering issues using a variant of this script: https://gist.github.com/erikarvstedt/62c38855e0092cd4adffa83cf751023b.

The script runs the nix-bitcoin test framework which starts bitcoind and a bunch of dependent services (CLN, elementsd, LND, joinmarket, mempool, fulcrum, btcpayserver, electrs) and checks whether they started up correctly.

[cbergqvist]

@jonasnick

What's the status of the disabling of JoinMarket?
https://gist.github.com/erikarvstedt/62c38855e0092cd4adffa83cf751023b#file-test-bitcoind-26-sh-L26-L27

[cbergqvist]

While going through the Testing Guide's v3 transaction section I noticed that if one attempts to do a testmempoolaccept just before the last submitpackage, the former would reject the input, but the latter would still allow it.

$ bcli testmempoolaccept "[\"$v3_parent_signed_tx\",\"$high_fee_child_signed_tx\"]"
[
  {
    "txid": "e40a9567347a3eace38f6e8c05d437c1c56460ad34ddeed724e595f776e16bff",
    "wtxid": "06551611e1c23087abd4f72d928c55ec69cf30cb19f5ffa7af00a87b8d77ad40",
    "allowed": false,
    "reject-reason": "txn-already-in-mempool"
  },
  {
    "txid": "66c8a8dfa2ba6995caae3dd1f109ea37408dd65f0a94cc2cd0872a4a5604a2a9",
    "wtxid": "0a67d798ad3f8be6875dfaa562cc1f2dd15b70e368c915cdca0e66e2a686c952"
  }
]

[glozow]

While going through the Testing Guide's v3 transaction section I noticed that if one attempts to do a testmempoolaccept just before the last submitpackage, the former would reject the input, but the latter would still allow it.

Yes, they have slightly different interfaces, for various reasons. For example

• Your example; testmempoolaccept errors when the tx is already there. I didn't change this when adding multi-tx functionality because it would be a change to an interface that had existed for a while. For submitpackage, I felt it would be much more useful to be tolerant of already-in-mempool transactions.
• We aren't really able to simulate package CPFPs in test-accepts at the moment (or at least it's a bit more difficult) so testmempoolaccept will fail on those

This isn't a bug, though I can see why it's confusing for people. Perhaps we can add more documentation about this.

[n1bor]

Testing on underpower host:
Intel(R) Atom(TM) CPU C2338 @ 1.74GHz
4Gb RAM
1Tb HDD (slow)

Had been running for 126 days with 500-600 incoming hosts and a full 36k tx mempool on 26.0rc2
After upgrade/restart was very slow at processing the mempool saved on disk - took 2 hours. Can not be sure this was not an issue in 26 also.

Is back up to 85 connections will monitor over next few days confirm performance is ok when up to a few hundred connections.

Edit 1: Now up to 205 ipv4, 140ipv6 and 26 onion - total 371. All looks fine. Waiting for mempool to refill will the test a restart to see how long take to read it in.

[0xB10C]

fwiw: #29785

[laanwj]

What's the status of the disabling of JoinMarket?

joinmarket (master) works with v27.0rc1, both sendpayment and the yield generator

[laanwj]

Closing this issue now that v27.0 final is tagged.