fac395e5eb ci: Bump ci/lint/Dockerfile (MarcoFalke)
fa6eb65167 test: Use python3.8 pow() (MarcoFalke)
88881cf7ac Bump python minimum version to 3.8 (MarcoFalke)
Pull request description:
There is no pressing reason to drop support for 3.7, however there are several maintenance issues:
* There is no supported operating system that ships 3.7 by default. (debian:buster is EOL and unmaintained to the extent that it doesn't run in the CI environment. See https://github.com/bitcoin/bitcoin/pull/27340#issuecomment-1484988445)
* Compiling python 3.7 from source is also unsupported on at least macos, according to https://github.com/bitcoin/bitcoin/pull/24017#issuecomment-1107820790
* Recent versions of lief require 3.8, see https://github.com/bitcoin/bitcoin/pull/27507#issuecomment-1517561645
Fix all maintenance issues by bumping the minimum.
ACKs for top commit:
RandyMcMillan:
ACK fac395e
fjahr:
ACK fac395e5eb
fanquake:
ACK fac395e5eb
Tree-SHA512: c198decdbbe29d186d73ea3f6549d8a38479383495d14a965a2f9211ce39637b43f13a4c2a5d3bf56e2d468be4bbe49b4ee8e8e19ec69936ff43ddf2b714c712
These routines look fancy, but do nothing more than converting between
byte objects of length 32 to/from integers in little endian byte order
and can be replaced by simple one-liners, using the int.{from,to}_bytes
methods (available since Python 3.2).
Also, move the burden of checking for a timeout to the client and
disable the timeout on the server. This should avoid intermittent issues
in slow tests (for example mining_getblocktemplate_longpoll.py, or
feature_dbcrash.py), or possibly when the server is running slow (for
example in valgrind). There shouldn't be any downside in tests caused
by a high rpcservertimeout.
This commit updates the code by replacing the RPC call used to
decode an address and retrieve its corresponding scriptpubkey
with the address_to_scriptpubkey function. address_to_scriptpubkey
function can now decode all addresses formats, which makes
it more efficient to use.
The COINBASE_MATURITY constant in blocktools.py is imported in wallet.py.
However, importing address_to_scriptpubkey to blocktools.py will
generate a circular import error. Since the method is related to
addresses, it is best to move it to address.py, which will also
fix the circular import error.
Update imports of address_to_scriptpubkey accordingly.
d178082996 test: add bech32 decoding support to address_to_scriptpubkey() (ismaelsadeeq)
aac8793c7a test: test_bech32_decode in address.py (ismaelsadeeq)
Pull request description:
[rpc_scantxoutset.py](e695d8536e/test/functional/rpc_scantxoutset.py (L26)) sendtodestination only sends to legacy addresses and scriptPubkeys because [wallet.py](e695d8536e/test/functional/test_framework/wallet.py (L415)) address_to_scriptpubkey does not support conversion of segwit address.
This update enables address_to_scriptpubkey to support the conversion of testnet segwit addresses to scriptPubkeys.
This change will enable [rpc_scantxoutset.py](e695d8536e/test/functional/rpc_scantxoutset.py (L22)) ScantxoutsetTest to have more test coverage by adding more sendtodestination calls with bech32 and bech32m testnet addresses, then test the bech32 and bech32m derivation subsets UTXO amount in [Test extended key derivation](e695d8536e/test/functional/rpc_scantxoutset.py (L84)).
I will add the test coverage in a subsequent Pull request.
ACKs for top commit:
josibake:
ACK d178082996
theStack:
ACK d178082996✔️
willcl-ark:
ACK d17808299
Tree-SHA512: 312c20ce192c648faf7dd178622700c9b871d755db56c246250e25508c3c19e7b02c0ae901dda11a1794629b9a9429c877168c05e1c4c1dbf41493316e30e7e9
Adds bech32_to_bytes() which can decode a bech32 address and return the
version as an `int` and the payload in bytes.
bech32_to_bytes() is used by the test_bech32_decode unit test to test
decoding of segwit addresses.
3dd2f6461b test: psbt: check non-witness UTXO removal for segwit v1 input (Sebastian Falbesoner)
dd78e3fa43 test: speedup rpc_psbt.py by whitelisting peers (immediate tx relay) (Sebastian Falbesoner)
e194e3e93d test: PSBT: eliminate magic numbers for global unsigned tx key (0) (Sebastian Falbesoner)
Pull request description:
This PR adds missing test coverage for dropping non-witness UTXOs from PSBTs for segwit v1+ inputs (see commit 103c6fd279). The formerly [disabled](4600479058) method `test_utxo_conversion` is re-enabled and adapted to spend a Taproot (`bech32m`) instead of a wrapped SegWit (`p2sh-segwit`) output. Note that in contrast to the original test, we have to add the non-witness UTXO manually here using the test framework's PSBT module, since the constructing node knows that the output is segwit v1 and hence doesn't add the non-witness UTXO in the first place (see also [BIP371]( https://github.com/bitcoin/bips/blob/master/bip-0371.mediawiki#user-content-UTXO_Types)).
I strongly assume that most wallets would behave the same as Bitcoin Core here and wouldn't create PSBTs with non-witness UTXOs for Taproot inputs, but it's still good to test everything works as expected if it's still done and that the non-witness UTXO is simply dropped in that case.
The first two commits contain a small refactor (magic number elimination in PSBT module) and test speedup of ~2-3x (using whitelisting peers / immediate tx relay).
ACKs for top commit:
achow101:
ACK 3dd2f6461b
instagibbs:
ACK 3dd2f6461b
Tree-SHA512: b8d7f7ea5d7d21def024b70dfca61991cc96a4193be8857018b4d7cf3ca1465d185619fd4a77623803d9da309aa489c53273e9b7683d970ce12e2399b5b50031
fa27cf4cc7 test: Default timeout factor to 4 under --valgrind (MarcoFalke)
Pull request description:
valgrind will incur a slowdown of at least 2, so increase the default timeout factor.
This should reduce the number of reported issues. See also https://github.com/bitcoin/bitcoin/issues/27112#issuecomment-1455762739
ACKs for top commit:
fanquake:
ACK fa27cf4cc7 - I still see at least one actual issue when running the functional tests under `--valgrind` (outside the CI system), but will follow up separately with that. Increasing the timeout here seems fine.
Tree-SHA512: 4467559a3bfd98f5735f300f6ed54c68f951191d65a2b1294d71d72cc5d0864964de562d5dfa0a4855fc541ccb269a538b7aeb3d408d2d012a5369513397c395
4d84eaec82 Raise PRNG seed log to INFO. (roconnor-blockstream)
Pull request description:
Some build infrastructure, such as Nix, will delete failed builds by default, keeping only the log (stdout/stderr) of the failed build.
For flaky tests, it would be very helpful to have the PRNG seed in the default log in order to redo the failed test.
By simply raising the PRNG seed logging to INFO, we can, by default, record the seed in the log of every build.
ACKs for top commit:
MarcoFalke:
lgtm ACK 4d84eaec82
theStack:
ACK 4d84eaec82
Tree-SHA512: 3ccb4a4e7639a3babc3b2a6456a6d0bffc090da34e4545b317f7bfbed4e9950d1b38ea5b2a90c37ccb49b3454bdeff03a6aaf86770b9c4dd14b26320aba50b94
61bb4e783b lint: enable E722 do not use bare except (Leonardo Lazzaro)
Pull request description:
Improve test code and enable E722 lint check.
If you want to catch all exceptions that signal program errors, use except Exception: (bare except is equivalent to except BaseException:).
Reference: https://peps.python.org/pep-0008/#programming-recommendations
ACKs for top commit:
MarcoFalke:
lgtm ACK 61bb4e783b
Tree-SHA512: c7497769d5745fa02c78a20f4a0e555d8d3996d64af6faf1ce28e22ac1d8be415b98e967294679007b7bda2a9fd04031a9d140b24201e00257ceadeb5c5d7665
Some build infrastructure, such as Nix, will delete failed builds by default, keeping only the log of the failed build.
For flaky tests, it would be very helpful to have the PRNG seed in the default log in order to redo the failed test.
By simply raising the PRNG seed logging to INFO, we can, by default, record the seed in the log of every build.
14302a4802 test: fix test abort for high timeout values (and `--timeout-factor 0`) (Sebastian Falbesoner)
Pull request description:
On master, the functional tests's option `--timeout-factor 0` (which according to the test docs and parameter description should disable the RPC timeouts) currently fails, same as high values like `--timeout-factor 999999`:
```
$ ./test/functional/wallet_basic.py --timeout-factor 0
2022-08-29T01:26:39.561000Z TestFramework (INFO): Initializing test directory /tmp/bitcoin_func_test_f24yxzp5
2022-08-29T01:26:40.262000Z TestFramework (ERROR): Assertion failed
Traceback (most recent call last):
File "/home/honey/bitcoin/test/functional/test_framework/test_framework.py", line 549, in start_nodes
node.wait_for_rpc_connection()
File "/home/honey/bitcoin/test/functional/test_framework/test_node.py", line 234, in wait_for_rpc_connection
rpc.getblockcount()
File "/home/honey/bitcoin/test/functional/test_framework/coverage.py", line 49, in __call__
return_val = self.auth_service_proxy_instance.__call__(*args, **kwargs)
File "/home/honey/bitcoin/test/functional/test_framework/authproxy.py", line 142, in __call__
response, status = self._request('POST', self.__url.path, postdata.encode('utf-8'))
File "/home/honey/bitcoin/test/functional/test_framework/authproxy.py", line 107, in _request
self.__conn.request(method, path, postdata, headers)
File "/usr/local/lib/python3.9/http/client.py", line 1285, in request
self._send_request(method, url, body, headers, encode_chunked)
File "/usr/local/lib/python3.9/http/client.py", line 1331, in _send_request
self.endheaders(body, encode_chunked=encode_chunked)
File "/usr/local/lib/python3.9/http/client.py", line 1280, in endheaders
self._send_output(message_body, encode_chunked=encode_chunked)
File "/usr/local/lib/python3.9/http/client.py", line 1040, in _send_output
self.send(msg)
File "/usr/local/lib/python3.9/http/client.py", line 980, in send
self.connect()
File "/usr/local/lib/python3.9/http/client.py", line 946, in connect
self.sock = self._create_connection(
File "/usr/local/lib/python3.9/socket.py", line 844, in create_connection
raise err
File "/usr/local/lib/python3.9/socket.py", line 832, in create_connection
sock.connect(sa)
OSError: [Errno 22] Invalid argument
```
This is caused by a high timeout value that Python's HTTP(S) client library can't cope with. Fix this by clamping down the connection's set timeout value in AuthProxy. The change can easily be tested by running an arbitrary test with `--timeout-factor 0` on master (should fail), on this PR (should pass) and on this PR with the clamping value increased by 1 (should fail).
// EDIT: The behaviour was observed on OpenBSD 7.1 and Python 3.9.12.
ACKs for top commit:
MarcoFalke:
lgtm ACK 14302a4802
Tree-SHA512: 6469e8ac699f1bb7dea11d5fb8b3ae54d895bb908570587c5631144cd41fe980ca0b1e6d0b7bfa07983307cba15fb26ae92e6766375672bf5be838d8e5422dbc
4159ccd031 test: Run feature_bip68_sequence.py with MiniWallet (Miles Liu)
fc0caaf4aa test: Add "include mempool" flag to MiniWallet rescan_utxos (Miles Liu)
d0a909ae54 test: Add "include immature coinbase" flag to MiniWallet get_utxos (Miles Liu)
e5b9127d9e test: Add signs P2TR and RAWSCRIPT to MiniWallet sign_tx (Miles Liu)
Pull request description:
This PR enables one more of the non-wallet functional tests (feature_bip68_sequence.py) to be run even when no wallet is compiled in by using the MiniWallet instead, as proposed in #20078.
ACKs for top commit:
achow101:
ACK 4159ccd031
MarcoFalke:
review ACK 4159ccd031🤸
Tree-SHA512: e891b189381e961c840b45fc30d058363707fd54c1c4bdc3d29623b03309981f1d3ebfac27e6aecf621bdbcd7e31754a3ef7c53f86346f7dd241c137e64c92bd
d6fc1d6a33 test: add coverage for dust mempool policy (`-dustrelayfee` setting) (Sebastian Falbesoner)
8a5dbe2879 test: add `CScript` method for checking for witness program (Sebastian Falbesoner)
Pull request description:
This PR adds missing test coverage for the `-dustrelayfee` setting, which specifies the fee-rate used to define dust. Output scripts for all common types that are treated as standard by default (P2PK, P2(W)PKH, P2(W)SH, P2TR, bare multisig, null data, unknown witness versions v2+) are created and then checked for dust-mempool-policy each via the `testmempoolaccept` RPC: a tx with an output's nValue equal to the dust threshold should be accepted, one with an nValue of just one 1 satoshi below that should be rejected with reason `dust`. This is repeatedly done for a fixed (but obviously somewhat arbitrary) list of different `-dustrelayfee` settings on a single node, including the default and zero (i.e. no dust limit) settings.
Note that the first commit introduces a necessary `CScript` helper method `IsWitnessProgram` (using PascalCase in Python is likely controversial; in this case the style for the already existing method `GetSigOpCount` was followed, which also refers to a method in the core `CScript` class).
Some historical information about dust, contributed by pablomartin4btc:
"The concept of dust was first introduced in https://github.com/bitcoin/bitcoin/pull/2577. This [commit](eb30d1a5b2) from https://github.com/bitcoin/bitcoin/pull/9380 introduced the -dustrelayfee option. Previous to that PR, the dust feerate was whatever -minrelaytxfee was set to."
ACKs for top commit:
LarryRuane:
ACK d6fc1d6a33
glozow:
ACK d6fc1d6a33
kouloumos:
ACK d6fc1d6a33
Tree-SHA512: 35ea2b2497dfb466395af5665bb217f7250aa7cab9dc43539a5658ab69a454e3623ff58fce7489fcc1105b37f8cb4840a93cec658c5df1de611732bc6439ccad
b2aa9e8528 Add release note for MIN_STANDARD_TX_NONWITNESS_SIZE relaxation (Greg Sanders)
8c5b3646b5 Relax MIN_STANDARD_TX_NONWITNESS_SIZE to 65 non-witness bytes (Greg Sanders)
Pull request description:
Since the original fix was set to be a "reasonable" transaction to reduce allocations and the true motivation later revealed, it makes sense to relax this check to something more principled.
There are more exotic transaction patterns that could take advantage of a relaxed requirement, such as 1 input, 1 output OP_RETURN to burn a utxo to fees for CPFP purposes when change isn't practical.
Two changes could be accomplished:
1) Anything not 64 bytes could be allowed
2) Anything above 64 bytes could be allowed
In the Great Consensus Cleanup, suggestion (2)
was proposed as a consensus change, and is the simpler of the two suggestions. It would not allow an "empty" OP_RETURN but would reduce the required padding from 22 bytes to 5.
The functional test is also modified to test the actual case
we care about: 64 bytes
Related mailing list discussions here:
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2022-October/020995.html
And a couple years earlier:
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2020-May/017883.html
ACKs for top commit:
achow101:
reACK b2aa9e8528
glozow:
reACK b2aa9e8528
pablomartin4btc:
re-ACK b2aa9e8528
jonatack:
ACK b2aa9e8528 with some suggestions
Tree-SHA512: c1ec1af9ddcf31b2272209a4f1ee0c5607399f8172e5a1dfd4604cf98bfb933810dd9369a5917ad122add003327c9fcf6ee26995de3aca41d5c42dba527991ad
Since the original fix was set to be a "reasonable" transaction
to reduce allocations and the true motivation later revealed,
it makes sense to relax this check to something more principled.
There are more exotic transaction patterns that could take advantage
of a relaxed requirement, such as 1 input, 1 output OP_RETURN to burn
a utxo to fees for CPFP purposes when change isn't practical.
Two changes could be accomplished:
1) Anything not 64 bytes could be allowed
2) Anything above 64 bytes could be allowed
In the Great Consensus Cleanup, suggestion (2) was the route taken.
It would not allow an "empty" OP_RETURN
but would reduce the required padding from 22 bytes to 5.
The functional test is also modified to test the actual case
we care about: 64 bytes
1647a11f39 tests: Reorder longer running tests in test_runner (Andrew Chow)
ff6c9fe027 tests: Whitelist test p2p connection in rpc_packages (Andrew Chow)
8c20796aac tests: Use waitfornewblock for work queue test in interface_rpc (Andrew Chow)
6c872d5e65 tests: Initialize sigops draining script with bytes in feature_taproot (Andrew Chow)
544cbf776c tests: Use batched RPC in feature_fee_estimation (Andrew Chow)
4ad7272f8b tests: reduce number of generated blocks for wallet_import_rescan (Andrew Chow)
Pull request description:
When configured with `--enable-debug`, many tests become dramatically slower. These slow downs are particularly noticed in tests that generate a lot of blocks in separate calls, make a lot of RPC calls, or send a lot of data from the test framework's P2P connection. This PR aims to improve the runtime of some of the slower tests and improve the overall runtime of the test runner. This has improved the runtime of the test runner from ~400s to ~140s on my computer.
The slowest test by far was `wallet_import_rescan.py`. This was taking ~320s. Most of that time was spent waiting for blocks to be mined and then synced to the other nodes. It was generating a new block for every new transaction it was creating in a setup loop. However it is not necessary to have one tx per block. By mining a block only every 10 txs, the runtime is improved to ~61s.
The second slowest test was `feature_fee_estimation.py`. This test spends most of its time waiting for RPCs to respond. I was able to improve its runtime by batching RPC requests. This has improved the runtime from ~201s to ~140s.
In `feature_taproot.py`, the test was constructing a Python `CScript` using a very large list of `OP_CHECKSIG`s. The constructor for the Python implementation of `CScript` was iterating this list in order to create a `bytes` from it even though a `bytes` could be created from it without iterating. By making the `bytes` before passing it into the constructor, we are able to improve this test's runtime from ~131s to ~106s.
Although `interface_rpc.py` was not typically a slow test, I found that it would occasionally have a super long runtime. It typically takes ~7s, but I have observed it taking >400s to run on occasion. This longer runtime occurs more often when `--enable-debug`. This long runtime was caused by the "exceeding work queue" test which is really just trying to trigger a race condition. In this test, it would create a few threads and try an RPC in a loop in the hopes that eventually one of the RPCs would be added to the work queue while another was processing. It used `getrpcinfo` for this, but this function is fairly fast. I believe what was happening was that with `--enable-debug`, all of the code for receiving the RPC would often take longer to run than the RPC itself, so the majority of the requests would succeed, until we got lucky after 10's of thousands of requests. By changing this to use a slow RPC, the race condition can be triggered more reliably, and much sooner as well. I've used `waitfornewblock` with a 500ms timeout. This improves the runtime to ~3s consistently.
The last test I've changed was `rpc_packages.py`. This test was one of the higher runtime variability tests. The main source of this variation appears to be waiting for the test node to relay a transaction to the test framework's P2P connection. By whitelisting that peer, the variability is reduced to nearly 0.
Lastly, I've reordered the tests in `test_runner.py` to account for the slower runtimes when configured with `--enable-debug`. Some of the slow tests I've looked at were listed as being fast which was causing overall `test_runner.py` runtime to be extended. This change makes the test runner's runtime be bounded by the slowest test (currently `feature_fee_estimation.py` with my usual config (`-j 60`).
ACKs for top commit:
willcl-ark:
ACK 1647a11
Tree-SHA512: 529e0da4bc51f12c78a40d6d70b3a492b97723c96a3526148c46943d923c118737b32d2aec23d246392e50ab48013891ef19fe6205bf538b61b70d4f16a203eb
564b580bf0 test: Introduce MIN_BLOCKS_TO_KEEP constant (Aurèle Oulès)
71d9a7c03b test: Wallet imports on pruned nodes (Aurèle Oulès)
e6906fcf9e rpc: Enable wallet import on pruned nodes (Aurèle Oulès)
Pull request description:
Reopens#16037
I have rebased the PR, addressed the comments of the original PR and added a functional test.
> Before this change importwallet fails if any block is pruned. This PR makes it possible to importwallet if all required blocks aren't pruned. This is possible because the dump format includes key timestamps.
For reviewers:
`python test/functional/wallet_pruning.py --nocleanup` will generate a large blockchain (~700MB) that can be used to manually test wallet imports on a pruned node. Node0 is not pruned, while node1 is.
ACKs for top commit:
kouloumos:
ACK 564b580bf0
achow101:
reACK 564b580bf0
furszy:
ACK 564b580
w0xlt:
ACK 564b580bf0
Tree-SHA512: b345a6c455fcb6581cdaa5f7a55d79e763a55cb08c81d66be5b12794985d79cd51b9b39bdcd0f7ba0a2a2643e9b2ddc49310ff03d16b430df2f74e990800eabf
feature_fee_estimation has a lot of loops that hit the RPC many times in
succession in order to setup scenarios. Using batched requests for these
can reduce the test's runtime without effecting the test's behavior.
