caac999ff0fc5c98fa438b7e96fe1232f6573fd5 refactor: remove dependence on AddrManTest (josibake)
f961c477b56737c546c275e4d86cecfa3f75d48c refactor: check Good() in tried_collisions test (josibake)
207f1c825c632c54af009516d376d392ea9106fa refactor: make AddrMan::Good return bool (josibake)
Pull request description:
Previously, the `addrman_tried_collisions` test behaved in the following way:
1. add an address to addrman
2. attempt to move the new address to the tried table (using `AddrMan.Good()`)
3. verify that `num_addrs` matched `size()` to check for collisions in the new table
`AddrMan.size()`, however, returns the number of unique address in addrman, regardless of whether they are in new or tried. This means the test would still pass for addresses where a collision did occur in the tried table. After 3 collisions in the tried table, there would eventually be a collision in the new table when trying to add a new address, which was then detected by checking `num_addrs - collisions == size()`.
While the collision in the new table was caused by a collision in the tried table, the test is misleading as it's not directly testing for collisions in the tried table and misses 3 collisions before identifying a collision in the new table.
### solution
To more directly test the tried table, I refactored `AddrMan::Good()` to return a boolean after successfully adding an address to the tried table. This makes the test much cleaner by first adding an address to new, calling `Good` to move it to the tried table, and checking if it was successful or not. It is worth noting there are other reasons, aside from collisions, which will cause `Good` to return false. That being said, this is an improvement over the previous testing methodology.
Additionally, having `Good()` return a boolean is useful outside of testing as it allows the caller to handle the case where `Good` is unable to move the entry to the tried table (e.g a063647413/src/rpc/net.cpp (L945)).
### followup
As a follow up to this PR, I plan to look at the following places `Good()` is called and see if it makes sense to handle the case where it is unable to add an entry to tried:
* a063647413/src/rpc/net.cpp (L945)
* a063647413/src/net.cpp (L2067)
* a063647413/src/net_processing.cpp (L2708)
ACKs for top commit:
jnewbery:
utACK caac999ff0
mzumsande:
Code review ACK caac999ff0fc5c98fa438b7e96fe1232f6573fd5
Tree-SHA512: f328896b1f095e8d2581fcdbddce46fc0491731a0440c6fff01081fa5696cfb896dbbe1d183eda2c100f19aa111e1f8b096ef93582197edc6b791de563a58f99
Currently restorewallet() logic is written in the RPC layer
and it can´t be reused by GUI. So it reimplements this in the
wallet and interface sections and then, GUI can access it.
eaf6be0114a6d7763767da9496907fe8a670ff9e [net processing] Do not request transaction relay from feeler connections (John Newbery)
0220b834b175dc8c45a2c60213474a72c0ef8193 [test] Add testing for outbound feeler connections (John Newbery)
Pull request description:
Feelers are short-lived connections used to test the viability of peers. The bitcoind node will periodically open feeler connections to addresses in its addrman, wait for a `version` message from the peer, and then close the connection.
Currently, we set `fRelay` to `1` in the `version` message for feeler connections, indicating that we want the peer to relay transactions to us. However, we close the connection immediately on receipt of the `version` message, and so never process any incoming transaction announcements. This PR changes that behaviour to instead set `fRelay` to `0` indicating that we do not wish to receive transaction announcements from the peer.
This PR also extends the `addconnection` RPC to allow creating outbound feeler connections from the node to the test framework, and a test to verify that the node sets `fRelay` to `0` in the `version` message to feeler connections.
ACKs for top commit:
naumenkogs:
ACK eaf6be0114a6d7763767da9496907fe8a670ff9e
MarcoFalke:
review ACK eaf6be0114a6d7763767da9496907fe8a670ff9e 🏃
Tree-SHA512: 1c56837dbd0a396fe404a5e39f7459864d15f666664d6b35ad109628b13158e077e417e586bf48946a23bd5cbe63716cb4bf22cdf8781b74dfce6047b87b465a
Rather than try to infer a collision by checking `AddrMan::size`,
check whether or not moving to the tried table was successful by
checking the output from `AddrMan::Good`
If AddrMan::Good is unable to add an entry
to tried (for a number of reasons), return false.
This makes it much easier and cleaner to directly
test for tried collisions. It also allows anyone
calling Good() to handle the case where adding an
address to tried is unsuccessful.
Update docs to doxygen style.
fa19bab90a3ccc2f76c20aa805292d6a9c5d8071 fuzz: Rework FillNode (MarcoFalke)
fae6e31df7c6df04f41fc8401e2a9781a4d75be7 refactor: Set fSuccessfullyConnected in FillNode (MarcoFalke)
fa3583f856e34b6c6134745da14f5206cf71fa3e fuzz: Avoid negative NodeId in ConsumeNode (MarcoFalke)
Pull request description:
Currently `FillNode` is a bit clumsy because it directly modifies memory of `CNode`. This gets in the way of moving that memory to `Peer`. Also, it isn't particularly consistent. See for example https://github.com/bitcoin/bitcoin/pull/21160#discussion_r739206139 .
Fix all issues by sending a `version`/`verack` in `FillNode` and let net_processing figure out the internal details.
ACKs for top commit:
jnewbery:
Strong concept ACK and light code review ACK fa19bab90a3ccc2f76c20aa805292d6a9c5d8071
Tree-SHA512: 33261d857c3fa6d5d39d742624009a29178ad5a15eb3fd062da741affa5a4854fd45ed20d59a6bba2fb068cf7b39cad6f95b2910be7cb6afdc27cd7917955b67
9544ab60ce212a29306f021c2cc6721c34485b54 doc: Improve ZMQ documentation (node01)
Pull request description:
This PR intends to clarify:
. when ZMQ notifications occur
. the message structure of each topic
Closes https://github.com/bitcoin/bitcoin/issues/23452#issue-1046579585
ACKs for top commit:
theStack:
ACK 9544ab60ce212a29306f021c2cc6721c34485b54
Tree-SHA512: 3b1314ad5eb8c359ffabd91ce9b47cf6cf8efa69be083a3bb5d865833ac67d7a88a8e012c4a4f59f2cd2a0e7f491e828897e85e9d01bae72ee83c4a6ad459f6e
1ef2c034523d0dfc36c84186d7a110cfc7de238b Add multisig tutorial (lsilva01)
Pull request description:
This PR adds a mutisig tutorial, as requested in https://github.com/bitcoin/bitcoin/issues/21278
Although there is already a brief explanation and a functional test about the multisig implemented in https://github.com/bitcoin/bitcoin/pull/22067, this tutorial proposes to use the signet (instead of regtest), bringing the reader closer to a real environment and explaining some functions in more detail.
I'm not sure if this format should be in this repository or on some wiki page. But as there is an open issue regarding this matter, that is my suggestion.
ACKs for top commit:
Sjors:
re-utACK 1ef2c034523d0dfc36c84186d7a110cfc7de238b
prayank23:
ACK 1ef2c03452
Tree-SHA512: 2c3f17a8c50e554f802029dceb28ab90a77021f135b8cbd77dca3879ba1f1a0eac6bda0afb90d1ff6b8116fb0628471687d3fb77bb255ef5d8b9590b775cbce9
fa26c5564429d3201f43850afaf1b4df2ed11ad9 wallet: Replace Assume with Assert where needed in coinselection (MarcoFalke)
Pull request description:
`Assume` should only be used when a failed check is recoverable. The checks here don't recover and would run into UB, so use `Assert` instead.
ACKs for top commit:
theStack:
Code-review ACK fa26c5564429d3201f43850afaf1b4df2ed11ad9
Tree-SHA512: 0cf9435f9ec44794022ce0274cba602aec95102ab73f4c8a93dae54ef4c0a594f6a81640477039719ddfb6f23b05f8ece3e4886ef7f8a725efff45685ac49d92
Ensure that blocks past the snapshot base block (i.e. the end of the
assumed-valid region of the chain) are not included in
setBlockIndexCandidates for the background validation chainstate. These
blocks, while fully validated and lacking the BLOCK_ASSUMED_VALID flag,
*rely* on blocks which are assumed-valid, and so shouldn't be added to
the IBD chainstate.
Co-authored-by: Russ Yanofsky <russ@yanofsky.org>
Avoid modifying the genesis block index entry during snapshot load. This
is because, in a future change that fixes LoadBlockIndex for UTXO
snapshots, we detect block index entries that are reliant on
assumed-valid ancestors and treat them specially.
Since the genesis block doesn't have BLOCK_VALID_SCRIPTS, it would be
erroneously marked BLOCK_ASSUMED_VALID during snapshot load if we didn't
skip it here. This would cause a "setBlockIndexCandidates() empty"
assertion to be tripped since all block index entries would be marked
assume-valid due to genesis, which is never re-validated.
There's probably no good reason to modify the genesis block index entry
during snapshot load anyway...
fa24a3df8796cbf4eeb35d950a4c848d605e5b22 rpc: Quote user supplied strings in error messages (MarcoFalke)
Pull request description:
I can't see a downside doing this and this fixes a fuzzing crash
Background:
This is a follow-up to commit 926fc2a0d4ff64cf2ff8e1dfa64eca2ebd24e090, which introduced the "starts_with-hack". Maybe an alternative to the hack would be to assign a unique error code to internal bugs? However, I think this can be done in an separate pull request and the changes here make sense even on their own.
ACKs for top commit:
fanquake:
ACK fa24a3df8796cbf4eeb35d950a4c848d605e5b22 - to fix the fuzzers.
Tree-SHA512: d998626406a64396a037a6d1fce22fce3dadb7567c2f9638e450ebe8fb8ae77d134e15dd02555326732208f698d77b0028bc62be9ceee9c43282b61fe95fccbd
e9440aeb5cad98fea9971f5126461e0a2b30ab54 build: use __SIZEOF_INT128__ for checking __int128 availability (fanquake)
Pull request description:
We already use this in the blockfilter code,
bf66e258a8/src/blockfilter.cpp (L34-L36)
so not sure we need to maintain two different ways of testing
for the same functionality. Consolidate on testing for `__SIZEOF_INT128__`,
which we already use, is supported by the compilers we care about, and is
also used by libsecp256k1.
ACKs for top commit:
sipa:
utACK e9440aeb5cad98fea9971f5126461e0a2b30ab54
Zero-1729:
crACK e9440aeb5cad98fea9971f5126461e0a2b30ab54
Tree-SHA512: 8aeef1734486a863b5091123bb5f9ba8868b1e2b4b35114586e3eb5862a38d4a1518ed069f37f41cb5e5ce2f6c87d95671996366d5ee990e0c90f268a8978ba3
41b9f7d062a01cf39b6dba02703213ffb1951a37 test: Use byte unit 'M' for -maxuploadtarget functional test (Douglas Chimento)
Pull request description:
ACKs for top commit:
shaavan:
ACK 41b9f7d062a01cf39b6dba02703213ffb1951a37
stratospher:
ACK 41b9f7d.
Tree-SHA512: 25b46347c671e8d6fd8878e7fee40e773bb03641e53e41e8a79a286fe4a0cf71c0c0986d6d7418fcb656c07f7216cc50a7ee4366f9213c32b01ae74326031f80
50209a42ad05e92ad0945fa104c65d2b1f2a2c1c validation, doc: remove TODO comment (Jon Atack)
8e37fa83934ad47a4a73c326213e2073dfc35e29 validation, log: improve logging in FlushSnapshotToDisk() (Jon Atack)
271252c0bda5779a255f23a73ceb6f722559f0d1 validation, log: extract FlushSnapshotToDisk() function (Jon Atack)
Pull request description:
Use the `LOG_TIME_MILLIS_WITH_CATEGORY_MSG_ONCE` macro to improve the logging of ChainstateManager snapshot persistance, log task start and completion separately and no longer manually track the duration, as suggested by Marco Falke in https://github.com/bitcoin/bitcoin/pull/22872#discussion_r715571280.
Extract the flushing into one function, which clarifies the logic, extends the improved logging to both flushing call sites, and allows logging the prefix `FlushSnapshotToDisk`, which is similar to `FlushStateToDisk`.
before
```
[snapshot] flushing coins cache (0 MB)... done (0.00ms)
[snapshot] flushing snapshot chainstate to disk
```
after
```
FlushSnapshotToDisk: flushing coins cache (0 MB) started
...
FlushSnapshotToDisk: completed (0.00ms)
FlushSnapshotToDisk: saving snapshot chainstate (0 MB) started
...
FlushSnapshotToDisk: completed (0.00ms)
```
The logging can be observed in the output of
```
./src/test/test_bitcoin -t validation_chainstate_tests -- DEBUG_LOG_OUT
```
Top commit has no ACKs.
Tree-SHA512: 5d954cd8c7455f8625152a43663a237f04717bb834aed62925a56e17c711fca6ccfc03783970b6b0bde44f64617d804b423a7048287c06ee816db36247acf272
We already use this in the blockfilter code, so not sure we need to maintain two
different ways of testing for the same functionality. Consolidate on testing
for __SIZEOF_INT128__, which we already use, is supported by the compilers we
care about, and is also used by libsecp256k1.
9600ea01450b0d39be90eb2971c1ac5c9b69a66e test: Add edge case of pruning up to index height (Martin Zumsande)
698c524698c33595a4d555eaa9e21bc19b4d3e93 index: Fix backwards search for bestblock (Martin Zumsande)
Pull request description:
This PR attempts to fix an intermittent Init issue encountered during the stress testing of #23289, which relates to the pruning-compatible filter reconstruction logic introduced in #15946.
The problem would occur when the node starts with `-txindex=1` but `ThreadSync` is interrupted after it sets `m_best_block_index` to Genesis, and before it gets do any further work.
In that case, during the next restart of the node, an Init error would be thrown because `BaseIndex::Init()` tries to backtrack from the tip to the last block which has been successfully indexed (here: Genesis), but the backtracking logic didn't work properly in this case:
The loop
`while (block_to_test && block->pprev && (block->pprev->nStatus & BLOCK_HAVE_DATA))` checks if a predecessor exists **before** performing the check `block_to_test == block` and then possbily setting `prune_violation = false`
If `block_to_test` and `block` are the Genesis block this check will not be reached because `block->pprev` does not exist.
To reproduce this bug on regtest:
1) start a node with a fresh datadir using `-txindex=1` (or any other index)
2) stop and restart without any index
3) mine a block
3) stop and restart again with the index enabled
->InitError `Error: txindex best block of the index goes beyond pruned data. (...)`
Fix this by requiring that we have the data for the block of the current iteration `block` (instead of requiring it for the predecessor `block->pprev`)
That way, the check for `block_to_test == block` is also reached when `block_to_test` is the Genesis block.
No longer requiring the data of `block->pprev` also means that we can now prune up to `m_best_block_index` height without requiring a reindex (one block more than before). I added this edge case to `feature_blockfilterindex_prune.py`, the new version should fail on master.
ACKs for top commit:
ryanofsky:
Partial code review ACK 9600ea01450b0d39be90eb2971c1ac5c9b69a66e for the code change, not the test changes. (Test changes are indirect and little over my head.) It seems obvious that previous code `prune_violation = true, while (block->pprev)` would incorrectly detect a prune violation at the genesis block, and the fix here make sense and looks correct.
Tree-SHA512: c717f372cee8fd49718b1b18bfe237aa6ba3ff4468588c10e1272d7a2ef3981d10af4e57de51dec295e2ca72d441bc6c2812f7990011a94d7f818775e3ff1a38
a9256dc3402a6e39ff2267efc892c0d3d41310b6 rpc: output all hash preimages in 'decodepsbt' (Antoine Poinsot)
4d6b5321a581548d9cdd6aea2dca56cfd96723ee psbt: implement hash preimages fields (Antoine Poinsot)
Pull request description:
This implements the [bip174 input fields for hash preimages](https://github.com/bitcoin/bips/blob/master/bip-0174.mediawiki).
One motivation is that we will need those once we implement signing support for Miniscript descriptors.
ACKs for top commit:
achow101:
ACK a9256dc3402a6e39ff2267efc892c0d3d41310b6
Sjors:
re-tACK a9256dc3402a6e39ff2267efc892c0d3d41310b6
w0xlt:
reACK a9256dc
Tree-SHA512: 145d39f7de86256d4174d063dcee217ea6f9ec7a138bbd5205941d17ca99dcccef0ace05f6d0d6a77dd863e3877b05e0752f2bc36ecd8c508e2c8adae2e03ae1
ef81c51f62e03cb53e65f1ced6120d87f1827c40 build: Use config_opts instead of passing a compiler flag directly (Hennadii Stepanov)
3b8f4e347b09594dcf88f4f2c0bb5afc1265ef51 build, qt: Drop unused CROSS_COMPILE variable for Android builds (Hennadii Stepanov)
e5a81e65d74382a1c8803e5c85ce7f0e25611a8c build, refactor: Re-use `qt_lib_suffix` variable (Hennadii Stepanov)
Pull request description:
This PR suggests small cleanups as follow ups of bitcoin/bitcoin#23489.
ACKs for top commit:
fanquake:
ACK ef81c51f62e03cb53e65f1ced6120d87f1827c40
Tree-SHA512: c834c4f76218c7d567223c49bb86404b02facf61bc9be9a66a984ddd2b2f5e6150ba97071c993ea1faf4da10c87e81df292b3628d4407afccbe9df379ac452d7
Since Qt 5.14.0 the CROSS_COMPILE variable always being overridden by
the Qt build system internally.
See upstream commit: c28b881c98fadcd3415370fad2525b558f6b03e4
It would make for sense for the TODO to be done in PR 17487
(or noted in the review feedback for a follow-up),
no need to continue maintaining the TODO in the codebase.
d5cab1a96d26e66d342fb5ec35c809bb82869d00 Add createmultisig and addmultisigaddress warnings release note (Samuel Dobson)
e46fc935aa6eb392ef9411dad310697aba601f6a Add warnings field to addmultisigaddress to warn about uncompressed keys (Samuel Dobson)
d1a9742623e2a8f3307c84c2df6993f62617a6f6 Add warnings field to createmultisig to warn about uncompressed keys (Samuel Dobson)
Pull request description:
Fixes#21368
Currently, if there are any uncompressed keys when calling `AddAndGetMultisigDestination`, it will just default to a legacy address regardless of the chosen `address_type`. Rather than keeping this silent behaviour which may be confusing to users, we explicitly add a `warnings` field which will warn the user why their address format is different.
ACKs for top commit:
achow101:
ACK d5cab1a96d26e66d342fb5ec35c809bb82869d00
Tree-SHA512: c2ac7f7689251bd4fcd8c26506f053921fbaf34c7a26a74e82ebc7f82cc0bd25407fd7954bf98365dcafa51fa45dcdbee6214320580ca69509690c3555e71cc0
