The previous diff touched most files in ./test/, so bump the headers to
avoid having to touch them again for a bump later.
-BEGIN VERIFY SCRIPT-
./contrib/devtools/copyright_header.py update ./test/
-END VERIFY SCRIPT-
facc352648 test: Implicitly sync after generate*, unless opted out (MarcoFalke)
Pull request description:
The most frequent failure in functional tests are intermittent races. Fixing such bugs is cumbersome because it involves:
* Noticing the failure
* Fetching and reading the log to determine the test case that failed
* Adding a `self.sync_all()` where it was forgotten
* Spamming out a pr and waiting for review, which is already sparse
Also, writing a linter to catch those is not possible, nor is review effective in finding these bugs prior to merge.
Fix all future intermittent races caused by a missing sync_block call by calling `sync_all` implicitly after each `generate*`, unless opted out. This ensures that the code is race-free (with regards to blocks) when the tests pass once, instead of our current approach where the code can never be guaranteed to be race-free.
There are some scripted-diff cleanups (see https://github.com/bitcoin/bitcoin/pull/22567), but they will be submitted in a follow-up to reduce the conflicts in this pull.
ACKs for top commit:
lsilva01:
tACK facc352 on Ubuntu 20.04
brunoerg:
tACK facc352648 on MacOS 11.6
Tree-SHA512: 046a40a066b4a3bd28a3077bd654fa8887442dd1f0ec6fd11671865809ef02376f126eb667a1320ebd67b6e372c78c00dbf8bd25d86ed86f1d9a25363103ed97
Because of floating point precision issues, not all of the rounding done
is always correct. To fix this, the fee calculation for
assert_fee_amount is changed to better reflect how CFeeRate::GetFee does
it.
First the feerate is converted to an int representing sat/kvb. Then this
is multiplied by the transaction size, divivided by 1000, and rounded up
to the nearest sat. The result is then converted back to BTC (divided by
1e8) and then rounded down to the nearest sat to avoid precision errors.
b658d7d5c5 test: update assert_fee_amount() in test_framework/util.py (Jon Atack)
Pull request description:
Follow-up to 42e1b5d979 (#12486).
- update call to `round()` with our utility function `satoshi_round()` to avoid intermittent test failures
- rename `fee_per_kB` to `feerate_BTC_kvB` for precision
- store division result in `feerate_BTC_vB`
Possibly resolves#19418.
ACKs for top commit:
meshcollider:
utACK b658d7d5c5
Tree-SHA512: f124ded98c913f98782dc047a85a05d3fdf5f0585041fa81129be562138f6261ec1bd9ee2af89729028277e75b591b0a7ad50244016c2b2fa935c6e400523183
- update call to round() with satoshi_round() to avoid intermittent test failures
- rename fee_per_kB to feerate_BTC_kvB for precision
- store division result in feerate_BTC_vB
This avoids having to remember to set it whenever mocktime is used with
peer connections. Also, it might help avoiding disconnects when
attaching a debugger to a running test.
Surprisingly easy to do. Almost all of the diff resolution was mechanically
* replacing boost::variant with std::variant
* replacing Optional with std::optional
* then replacing `nullopt` with `std::nullopt`
* updating the RPC functions for the new RPCArg::Default type
* update the tests/ directory to make new (since 22) tests use arrays for
createrawtransaction outputs
* other ad-hoc changes to function parameters etc (not too many of these)
I had to "really" change the code in PrecomputePSBTData, which was introduced
in 22.0 and affected by PSET, but this function was like 8 lines long so it
was easy.
Reviewing the diff may be a bit difficult because of the mix of mechanical
changes and ad-hoc things. Probably the most straightforward thing to do
is to redo the merge, `sed -i` to fix the boost::variant and Optional stuff,
then diff the remaining conflicts against this commit.
TODO: grep for `blindpsbt` and you will see that this RPC is still referenced
in documentation and help text even though it was deleted. Need to fix this
in 0.21 in a separate PR.
8ca51af1ec test: Disable automatic connections by default (Martin Zumsande)
Pull request description:
A node normally doesn't make automatic connections to peers in the functional tests because neither DNS seeds nor hardcoded peers are available on regtest. However, when random entries are inserted into addrman as part of a functional test (e.g. while testing addr relay), `ThreadOpenConnections` will periodically try to connect to them, resulting in log entries such as:
`[opencon] [net.cpp:400] [ConnectNode] trying connection 18.166.1.1:8333 lastseen=0.0hrs`
I don't think it's desirable that functional tests try to connect to random computers on the internet, aside from the possibility that at some point in time someone out there might actually answer in a way to ruin a test.
This PR fixes this problem by disabling `ThreadOpenConnections` by adding `-connect=0` to the default args, and adding exceptions only when needed for the test to pass.
ACKs for top commit:
tryphe:
Concept ACK, light code review ACK 8ca51af1ec
Tree-SHA512: bcfb2de610e6c35a97a2bd7ad6267e968b1ac7529638d99276993cd5bc93ce9919d54e22d6dc84e1b02ecd626ab6554e201693552ea065c29794eece38c43f7d
This prevents the node from trying to connect to random IPs on the internet
while running the functional tests. Exceptions are added when required for
the test to pass.
This PR eliminates "strange regtest=0 behavior" in a test which had forced
us to disable the test for Elements. Can re-enable now :)
I also removed the `chain_in_args` parameter to `TestNode`, which Steven added
in https://github.com/ElementsProject/elements/pull/533 (which itself replaces
unconditionally adding chain={} on the command-line, which was added in #458).
These were added in the 0.17 rebase to deal with the job of starting bitcoind,
which then did not support the `chain=` command-line arg as well as elementsd,
which back then required this command-line arg.
This was causing some issues with the "check -acceptnonstdtxn doesn't work on
mainnet" test because it would add -chain=elementsregtest to the command-line
of a daemon that was supposed to be connecting to mainnet/liquidv1. It is
possible to override this behavior, but since 0.20+ versions of elementsd and
bitcoind have essentially the same support for chain= options, it seemed
cleaner to just eliminate the diff.
Since we want tests to run quickly, and since tests do a lot more db
operations than expected we expect to see in actual usage, we disable
sqlite's syncing behavior to make db operations run much faster. This
syncing behavior is necessary for normal operation as it helps guarantee
that data won't become lost or corrupted, but in tests, we don't care
about that.
1) add a new sane "address" field (for outputs that have an
identifiable address, which doesn't include bare multisig)
2) with -deprecatedrpc: leave "reqSigs" and "addresses" intact
(with all weird/wrong behavior they have now)
3) without -deprecatedrpc: drop "reqSigs" and "addresses" entirely,
always.
