cfcb9b1ecf test: wallet, coverage for concurrent db transactions (furszy)
548ecd1155 tests: Test for concurrent writes with db tx (Ava Chow)
395bcd2454 sqlite: Ensure that only one SQLiteBatch is writing to db at a time (Ava Chow)
Pull request description:
The way that we have configured SQLite to run means that only one database transaction can be open at a time. Typically, each individual read and write operation will be its own transaction that is opened and committed automatically by SQLite. However, sometimes we want these operations to be batched into a multi-statement transaction, so `SQLiteBatch::TxnBegin`, `SQLiteBatch::TxnCommit`, and `SQLiteBatch::TxnAbort` are used to manage the transaction of the database.
However, once a db transaction is begun with one `SQLiteBatch`, any operations performed by another `SQLiteBatch` will also occur within the same transaction. Furthermore, those other `SQLiteBatch`s will not be expecting a transaction to be active, and will abort it once the `SQLiteBatch` is destructed. This is problematic as it will prevent some data from being written, and also cause the `SQLiteBatch` that opened the transaction in the first place to be in an unexpected state and throw an error.
To avoid this situation, we need to prevent the multiple batches from writing at the same time. To do so, I've implemented added a `CSemaphore` within `SQLiteDatabase` which will be used by any `SQLiteBatch` trying to do a write operation. `wait()` is called by `TxnBegin`, and at the beginning of `WriteKey`, `EraseKey`, and `ErasePrefix`. `post()` is called in `TxnCommit`, `TxnAbort` and at the end of `WriteKey`, `EraseKey`, and `ErasePrefix`. To avoid deadlocking on ` TxnBegin()` followed by a `WriteKey()`, `SQLiteBatch will now also track whether a transaction is in progress so that it knows whether to use the semaphore.
This issue is not a problem for BDB wallets since BDB uses WAL and provides transaction objects that must be used if an operation is to occur within a transaction. Specifically, we either pass a transaction pointer, or a nullptr, to all BDB operations, and this allows for concurrent transactions so it doesn't have this problem.
Fixes#29110
ACKs for top commit:
josibake:
ACK cfcb9b1ecf
furszy:
ACK cfcb9b1ecf
ryanofsky:
Code review ACK cfcb9b1ecf. This looks great and I think it is ready for merge. Just holding off because josibake seemed ready to review https://github.com/bitcoin/bitcoin/pull/29112#issuecomment-1930372190 and might have more feedback.
Tree-SHA512: 2dd5a8e76df52451a40e0b8a87c7139d68a0d8e1bf2ebc79168cc313e192dab87cfa4270ff17fea4f7b370060d3bc9b5d294d50f7e07994d9b5a69b40397c927
a17fd33edd GUI: OptionsDialog: Replace verbose two-option font selector with simple combobox with Custom... choice (Luke Dashjr)
98e9ac5199 GUI: Use FontChoice type in OptionsModel settings abstraction (Luke Dashjr)
3a6757eed9 GUI: Load custom FontForMoney from QSettings (Luke Dashjr)
49eb97eff9 GUI: Add possibility for an explicit QFont for FontForMoney in OptionsModel (Luke Dashjr)
f2dfde80b8 GUI: Move "embedded font or not" decision into new OptionsModel::getFontForMoney method (Luke Dashjr)
Pull request description:
This replaces the overly-verbose radio-button font setting (which only allows embedded or autodetected from system) with a simple combobox providing both existing options as well as a custom option to allow the user to select any font of their choice/style.
ACKs for top commit:
pablomartin4btc:
tested ACK a17fd33edd
hebasto:
ACK a17fd33edd, I have reviewed the code and tested it on Ubuntu 22.04. This is a UX improvement. https://github.com/bitcoin-core/gui/pull/497#issuecomment-1341222673 might be addressed in a follow-up.
Tree-SHA512: 2f0a8bc1242a374c4b7dc6e34014400428b6d36063fa0b01c9f62a8bd6078adfbbca93d95c87e4ccb580d982fe10173e1d9a28bcec586591dd3f966c7b90fc5d
facafa90f7 test: Fix CPartialMerkleTree.nTransactions signedness (MarcoFalke)
Pull request description:
It is unsigned in Bitcoin Core, so the tests should match it:
aa9231fafe/src/merkleblock.h (L59)
Large positive values, or "negative" values, are rejected anyway, but it still seems fine to fix this.
The bug was introduced when the code was written in d280617bf5.
(Lowercase `i` means signed, see https://docs.python.org/3/library/struct.html#format-characters)
ACKs for top commit:
theStack:
LGTM ACK facafa90f7
Empact:
ACK facafa90f7
Tree-SHA512: 35ac11bb5382dffe132bfae6097efc343ef6c06b1b4b1545130ca27b228ca6894679004862fee921b095172abaddbef5972c24d9bc195ce970f35643bd4a0f09
b14298c5bc fuzz: remove unused `args` and `context` from `FuzzedWallet` (brunoerg)
Pull request description:
`ArgsManager args` and `WalletContext context` were previously used to create the wallet into `FuzzedWallet`. After fa15861763, they are not used anymore. This PR removes them.
ACKs for top commit:
maflcko:
lgtm ACK b14298c5bc
epiccurious:
utACK b14298c5bc
ryanofsky:
Code review ACK b14298c5bc
Tree-SHA512: 164e6a66ba05e11176a0cf68db6257f0ac07459cf7aa01ec4302b303c156c205a68128373a0b8daba0a6dfbff990af7fa14465a6341a296312fb20ea778c7a8c
e2ad343f69 wallet: remove unused `SignatureData` instances in spkm's `FillPSBT` methods (Sebastian Falbesoner)
Pull request description:
These are filled with signature data from a PSBT input, but not used anywhere after, hence they can be removed. Note that the same code is in the `SignPSBTInput` function where the `sigdata` result is indeed used.
ACKs for top commit:
achow101:
ACK e2ad343f69
brunoerg:
crACK e2ad343f69
Tree-SHA512: f0cabcc000bcea6bc7d7ec9d3be2e2a8accbdbffbe35252250ea2305b65a5813fde2b8096fbdd2c7cccdf417ea285183dc325fc2d210d88bce62978ce642930e
2bb25ce502 wallet: remove unused 'accept_no_keys' arg from decryption process (furszy)
Pull request description:
Found it while reviewing other PR. Couldn't contain myself from cleaning it up.
The wallet decryption process (`CheckDecryptionKey()` and `Unlock()`)
contains an arg 'accept_no_keys,' introduced in #13926, that has
never been used.
Additionally, this also removes the unimplemented `SplitWalletPath`
function.
ACKs for top commit:
delta1:
ACK 2bb25ce502
epiccurious:
utACK 2bb25ce502.
achow101:
ACK 2bb25ce502
theStack:
Code-review ACK 2bb25ce502
Tree-SHA512: e0537c994be19ca0032551d8a64cf1755c8997e04d21ee0522b31de26ad90b9eb02a8b415448257b60bced484b9d2a23b37586e12dc5ff6e35bdd8ff2165c6bf
There are occasions where a multi-statement tx is begun in one batch,
and a second batch is created which does a normal write (without a
multi-statement tx). These should not conflict with each other and all
of the data should end up being written to disk.
A SQLiteBatch need to wait for any other batch to finish writing before
it can begin writing, otherwise db txn state may be incorrectly
modified. To enforce this, each SQLiteDatabase has a semaphore which
acts as a lock and is acquired by a batch when it begins a write, erase,
or a transaction, and is released by it when it is done.
To avoid deadlocking on itself for writing during a transaction,
SQLiteBatch also keeps track of whether it has begun a transaction.
e7fd70f4b6 [test] make v2transport arg in addconnection mandatory and few cleanups (stratospher)
Pull request description:
- make `v2transport` argument in `addconnection` regression-testing only RPC mandatory. https://github.com/bitcoin/bitcoin/pull/24748#discussion_r1470738750
- previously it was an optional arg with default `false` value.
- only place this RPC is used is in the [functional tests](11b436a66a/test/functional/test_framework/test_node.py (L742)) where we always pass the appropriate `v2transport` option to the RPC anyways. (and that too just for python dummy peer(`P2PInterface`) and bitcoind(`TestNode`) interactions)
- rename `v2_handshake()` to `_on_data_v2_handshake()` https://github.com/bitcoin/bitcoin/pull/24748#discussion_r1466958424
- more compact return statement in `wait_for_reconnect()` https://github.com/bitcoin/bitcoin/pull/24748#discussion_r1466979708
- assertion to check that empty version packets are received from `TestNode`.
ACKs for top commit:
glozow:
ACK e7fd70f4b6
theStack:
Code-review ACK e7fd70f4b6
mzumsande:
Code Review ACK e7fd70f4b6
Tree-SHA512: e66e29baccd91e1e4398b91f7d45c5fc7c2841d77d8a6178734586017bf2be63496721649da91848dec71da605ee31664352407d5bb896e624cc693767c61a1f
c340503b67 test: p2p: adhere to typical VERSION message protocol flow (Sebastian Falbesoner)
7ddfc28309 test: p2p: process post-v2-handshake data immediately (Sebastian Falbesoner)
b198b9c2ce test: p2p: introduce helper for sending prepared VERSION message (Sebastian Falbesoner)
Pull request description:
This PR addresses a quirk in the test framework's p2p implementation regarding the version handshake protocol:
Currently, the VERSION message is sent immediately after an inbound connection (i.e. TestNode outbound connection) is made. This doesn't follow the usual protocol flow where the initiator sends a version first, the responder processes that and only then responds with its own version message. Change that accordingly by only sending immediate VERSION message for outbound connections (or after v2 handshake for v2 connections, respectively), and sending out VERSION message as response for incoming VERSION messages (i.e. in the function `on_version`) for inbound connections.
I first stumbled upon this issue through reading comment https://mirror.b10c.me/bitcoin-bitcoin/24748/#discussion_r1465420112 (see last paragraph) and recently again in the course of working on a v2-followup for #29279, where this causes issues for TestNode outbound connections that disconnect *before* sending out their own version message.
Note that these changes lead to slightly more code in some functional tests that override the `on_version` method, as the version reply has to be sent explicitly now, but I think is less confusing and reflects better what is actually happening.
ACKs for top commit:
epiccurious:
utACK c340503b67
stratospher:
tested ACK c340503b67. very useful to have since we'd want real node behaviour!
mzumsande:
ACK c340503b67
sr-gi:
tACK c340503b67
Tree-SHA512: 63eac287d3e1c87a01852bfd9f0530363354bbb642280298673b9c8817056356373adf348955c4e92af95c7c6efa8cc515cee2892e9f077bfbe1bce8e97ad082
c819a83b4d Don't use scientific notation in log messages (Kristaps Kaupe)
Pull request description:
Don't see any benefits here, only harder to read for most of the users.
Before:
```
2024-01-16T13:11:36Z Dumped mempool: 8.165e-06s to copy, 0.00224268s to dump
```
After:
```
2024-01-16T13:11:36Z Dumped mempool: 0.000s to copy, 0.002s to dump
```
ACKs for top commit:
kristapsk:
> > > > lgtm ACK [c819a83](c819a83b4d). can you update the PR description?
glozow:
lgtm ACK c819a83b4d. can you update the PR description?
Tree-SHA512: 0972e0a05934e1b014fdeca0c235065aa017ba9abf74b3018f514e4d8022ef02b7f042a07d3675144b51449492468aea6b5b0183233ad7f1bab887d18e3d06af
fa5cd66f0a test: Assumeutxo with more than just coinbase transactions (MarcoFalke)
Pull request description:
Currently the AU tests only check that loading a txout set with only coinbase outputs works.
Fix that by adding other transactions.
ACKs for top commit:
jamesob:
ACK fa5cd66f0a
glozow:
concept ACK fa5cd66f0a
Tree-SHA512: e090c41f73490ad72e36c478405bfd0716d46fbf5a131415095999da6503094a86689a179a84addae3562b760df64cdb67488f81692178c8ca8bf771b1e931ff
The wallet decryption process (CheckDecryptionKey() and Unlock())
contains an arg 'accept_no_keys,' introduced in #13926, that has
never been used.
Additionally, this also removes the unimplemented SplitWalletPath
function.
4da76ca247 test: Test migration of tx with both spendable and watchonly (Ava Chow)
c62a8d03a8 wallet: Keep txs that belong to both watchonly and migrated wallets (Ava Chow)
71cb28ea8c test: Make sure that migration test does not rescan on reloading (Ava Chow)
78ba0e6748 wallet: Reload the wallet if migration exited early (Ava Chow)
9332c7edda wallet: Write bestblock to watchonly and solvable wallets (Ava Chow)
Pull request description:
A transaction does not necessarily have to belong to either the migrated wallet (with the private keys) and the watchonly wallet (with watchonly things), it could have multiple outputs with each isminetype. So we should be putting such transactions in one or the other wallet, but rather putting it in both.
I've added a test for this behavior, however the test also revealed a few other issues. Notably, it revealed that `migratewallet` would have the watchonly wallet rescan from genesis when it is reloaded at the end of migration. This could be a cause for migration appearing to be very slow. This is resolved by first writing best block records to the watchonly and solvable wallets, as well as updating the test to make sure that rescans don't happen.
The change to avoid rescans also found an issue where some of our early exits would result in unloading the wallet even though nothing happened. So there is also a commit to reload the wallet for such early exits.
ACKs for top commit:
ryanofsky:
Code review ACK 4da76ca247. This looks great. The code is actually cleaner than before, two bugs are fixed, and the test checking for rescanning is pretty clever and broadens test coverage.
furszy:
Code review ACK 4da76ca2
Tree-SHA512: 5fc210cff16ca6720d7b2d0616d7e3f295c974147854abc704cf99a3bfaad17572ada084859e7a1b1ca94da647ad130303219678f429b7995f85e040236db35c
3904123da9 tests: Test that descriptors flag is set for migrated blank wallets (Ava Chow)
072d506240 wallet: Make sure that the descriptors flag is set for blank wallets (Ava Chow)
Pull request description:
While rebasing #28710 after #28976 was merged, I realized that although blank wallets were being moved to sqlite, `WALLET_FLAG_DESCRIPTORS` was not being set so those blank wallets would still continue to be treated as legacy wallets.
To fix that, just set the descriptor flags for blank wallets. Also added a test to catch this.
ACKs for top commit:
epiccurious:
Tested ACK 3904123da9.
delta1:
tested ACK 3904123da9
ryanofsky:
Code review ACK 3904123da9
murchandamus:
code review ACK 3904123da9
Tree-SHA512: 9f6fe9c1899ca387ab909b1bb6956cd6bc5acbf941686ddc6c061f9b1ceec2cc9d009ff472486fc86e963f6068f0e2f1ae96282e7c630193797a9734c4f424ab
It is possible for a transaction that has an output that belongs to the
mgirated wallet, and another output that belongs to the watchonly
wallet. Such transaction should appear in both wallets during migration.
We want to make sure that all of the transactions are being copied to
the watchonly and solvable wallets as expected. The automatic rescanning
behavior can cause us to pass a test by finding the transaction
on loading rather than having it be copied as expected.
Migration will unload loaded wallets prior to beginning. It will then
perform some checks which may exit early. Such unloaded wallets should
be reloaded prior to exiting.
When migrating, we should also be writing the bestblock record to the
watchonly and solvable wallets to avoid rescanning on the reload as that
can be slow.
25dc87e6f8 libconsensus: deprecate (Cory Fields)
Pull request description:
This library has existed for nearly 10 years with very little known uptake or impact. It has become a maintenance burden. In several cases it dictates our code/library structure (for example necessitating LIBBITCOIN_CRYPTO_BASE), as well as build-system procedures (building multiple copies of object files especially for the lib).
Several discussions have arisen wrt migrating it to CMake and it has become difficult to justify adding more complexity for a library that is virtually unused anyway.
See for example the discussions:
https://github.com/hebasto/bitcoin/pull/41https://github.com/bitcoin/bitcoin/pull/29123
And here: https://github.com/bitcoin/bitcoin/pull/29180
Where it is pointed out that the libbitcoinconsensus functions are slower than those the internal bitcoind equivalents due to the missing sha2 implementations.
Instead, we (fanquake, hebasto, TheCharlatan, and I) propose simply not migrating it to CMake and letting it end with v27. Any remaining use-cases could be handled in the future by libbitcoinkernel.
If there are any users currently using libbitcoinconsensus, please chime in with your use-case!
Edit: Corrected final release to be v27.
ACKs for top commit:
TheCharlatan:
ACK 25dc87e6f8
fanquake:
ACK 25dc87e6f8 - this library has very little, if any impactful real world usage. It has been entirely broken (on various platforms) for long periods of its existence, where nobody even noticed. Pruning this out to save porting, and starting anew with the kernel, is the sane thing to do.
Tree-SHA512: baff2b3c4f76f520c96021035f751fdcb51bedf00e767660249e92a7bc7c5c176786bcf2c4cfe2d2351c200f932b39eb886bcfb22fbec824a41617590d6a1638
fad74bbbd0 refactor: Mark prevector iterator with std::contiguous_iterator_tag (MarcoFalke)
fab8a01048 refactor: Fix binary operator+ for prevector iterators (MarcoFalke)
fa44a60b2b refactor: Fix constness for prevector iterators (MarcoFalke)
facaa66b49 refactor: Add missing default constructor to prevector iterators (MarcoFalke)
Pull request description:
Currently prevector iterators have many issues:
* Forward iterators (and stronger) must be default constructible (https://eel.is/c++draft/forward.iterators#1.2). Otherwise, some functions can not be instantiated, like `std::minmax_element`.
* Various `const` issues with random access iterators. For example, a `const iterator` is different from a `const_iterator`, because the first one holds a mutable reference and must also return it without `const`. Also, `operator+` must be callable regardless of the iterator object's `const`-ness.
* When adding an offset to random access iterators, both `x+n` and `n+x` must be specified, see https://eel.is/c++draft/random.access.iterators#tab:randomaccessiterator
Fix all issues.
Also, upgrade the `std::random_access_iterator_tag` (C++17) to `std::contiguous_iterator_tag` (C++20)
ACKs for top commit:
TheCharlatan:
ACK fad74bbbd0
stickies-v:
ACK fad74bbbd0
willcl-ark:
ACK fad74bbbd0
Tree-SHA512: b1ca778a31602af94b323b8feaf993833ec78be09f1d438a68335485a4ba97f52125fdd977ffb9541b89f8d45be0105076aa07b5726936133519aae832556e0b
The test framework's p2p implementation currently sends out it's VERSION
message immediately after an inbound connection (i.e. TestNode outbound
connection) is made. This doesn't follow the usual protocol flow where
the initiator sends a version first, and the responders processes that
and only then responds with its own version message. Change that
accordingly by only sending immediate VERSION message for outbound
connections (or after v2 handshake for v2 connections, respectively),
and sending out VERSION messages as response for incoming VERSION
messages (i.e. in the function `on_version`) for inbound connections.
Note that some of the overruled `on_version` methods in functional tests
needed to be changed to send the version explicitly.
In the course of executing the asyncio data reception callback during a
v2 handshake, it's possible that the receive buffer already contains
data for after the handshake (usually a VERSION message for inbound
connections).
If we don't process that data immediately, we would do so after the next
message is received, but with the adapted protocol flow introduced in
the next commit, there is no next message, as the TestNode wouldn't
continue until we send back our own version in `on_version`. Fix this by
calling `self._on_data` immediately if there's data left in the receive
buffer after a completed v2 handshake.
This deduplicates code for sending out the VERSION message
(if available and not sent yet), currently used at three
different places:
1) in the `connection_made` asyncio callback
(for v1 connections that are not v2 reconnects)
2) at the end of `v2_handshake`, if the v2 handshake succeeded
3) in the `on_version` callback, if a reconnection with v1 happens
b851c5385d fuzz: extend ConsumeNetAddr() to return I2P and CJDNS addresses (Vasil Dimov)
Pull request description:
In the process of doing so, refactor `ConsumeNetAddr()` to generate the addresses from IPv4, IPv6, Tor, I2P and CJDNS networks in the same way - by preparing some random stream and deserializing from it. Similar code was already found in `RandAddr()`.
ACKs for top commit:
achow101:
ACK b851c5385d
mzumsande:
ACK b851c5385d
brunoerg:
utACK b851c5385d
Tree-SHA512: 9905acff0e996f30ddac0c14e5ee9e1db926c7751472c06d6441111304242b563f7c942b162b209d80e8fb65a97249792eef9ae0a96100419565bf7f59f59676
9642aefb81 test: fix intermittent failure in p2p_v2_earlykeyresponse (Martin Zumsande)
Pull request description:
The test fails intermittently, see https://cirrus-ci.com/task/6403578080788480?logs=ci#L3521 and https://github.com/bitcoin/bitcoin/pull/24748#issuecomment-1916996716.
I think it's because of a race between the python NetworkThread and the actual
test, which will both call `initiate_v2_handshake`. I could reproduce it by adding a sleep into `initiate_v2_handshake` after the line `self.sent_garbage = random.randbytes(garbage_len)`.
Fix this by waiting for the first `initiate_v2_handshake` to have finished before calling it a second time.
ACKs for top commit:
stratospher:
tested ACK 9642aef.
achow101:
ACK 9642aefb81
theStack:
Tested ACK 9642aefb81
Tree-SHA512: f728bbceaf816ddefeee4957494ccb608ad4fc912cb5cbf5f2acf09836df969c4e8fa2bb441aadb94fa39b3ffbb005d4132e7b6a5a98d80811810d8bd1d624e3
987a1b51ee init: settings, do not load auto-generated warning msg (furszy)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/pull/29144#issuecomment-1907071391.
The settings warning message is meant to be used only to discourage users from
modifying the file manually. Therefore, there is no need to keep it in memory.
ACKs for top commit:
achow101:
ACK 987a1b51ee
ryanofsky:
Code review ACK 987a1b51ee. Seems like a clean, simple fix
Tree-SHA512: 3f2bdcf4b4a9cadb396dcff9b43155211eeed018527a07356970a341d139ad18edbd1a4d369377c8907b8ec1f19ee2ab8aacf85a887379e6d57a8a6db2403d51
c11c404281 tests: Test migration of blank wallets (Andrew Chow)
563b2a60d6 wallet: Better error message when missing LegacySPKM during migration (Andrew Chow)
b1d2c771d4 wallet: Check for descriptors flag before migration (Andrew Chow)
8c127ff1ed wallet: Skip key and script migration for blank wallets (Andrew Chow)
Pull request description:
Blank wallets (wallets without any keys are scripts) are being detected as already being descriptor wallets even though they are not. This is because the check for whether a wallet is already a descriptor wallet uses the presence of a `LegacyScriptPubKeyMan` which is only setup when keys or scripts are found. This PR resolves this issue by checking for the descriptor wallet flag instead and subsequently skipping the keys and scripts part of migration for blank wallets.
Fixes the issue mentioned in https://github.com/bitcoin/bitcoin/pull/28868#issuecomment-1809641110
ACKs for top commit:
furszy:
reACK c11c404281. CI failure is unrelated.
ryanofsky:
Code review ACK c11c404281
Tree-SHA512: 2466fdf1542eb8489c841253191f85dc88365493f0bb3395b67dee3e43709a9993c68b9d7623657b54b779adbe68fc81962d60efef4802c5d461f154167af7f4
ff9039f6ea Remove GetAdjustedTime (dergoegge)
Pull request description:
This picks up parts of #25908.
The use of adjusted time is removed from validation code while the warning to users if their clock is out of sync with the rest of the network remains.
ACKs for top commit:
naumenkogs:
ACK ff9039f6ea
achow101:
ACK ff9039f6ea
maflcko:
lgtm ACK ff9039f6ea🤽
stickies-v:
ACK ff9039f6ea
Tree-SHA512: d1f6b9445c236915503fd2ea828f0d3b92285a5dbc677b168453276115e349972edbad37194d8becd9136d8e7219b576af64ec51c72bdb1923e57e405c0483fc
0bef1042ce net: enable v2transport by default (Pieter Wuille)
Pull request description:
This enables BIP324's v2 transport by default (see #27634):
* Inbound connections will auto-sense whether v1 or v2 is in use.
* Automatic outbound connections will use v2 if `NODE_P2P_V2` was set in addr gossip, but retry with v1 if met with immediate failure.
* Manual outbound connections will default to v2, but retry with v1 if met with immediate failure.
It remains possible to run with `-v2transport=0` to disable all of these, and make all outbound and inbound connections v1. It also remains possible to specify the `v2transport` argument to the `addnode` RPC as `false`, to disable attempting a v2 connection for that particular added node.
ACKs for top commit:
stratospher:
ACK 0bef104.
josibake:
reACK 0bef1042ce
achow101:
ACK 0bef1042ce
naumenkogs:
ACK 0bef1042ce
theStack:
ACK 0bef1042ce
willcl-ark:
crACK 0bef1042ce
BrandonOdiwuor:
utACK 0bef1042ce
pablomartin4btc:
re ACK 0bef1042ce
kristapsk:
utACK 0bef1042ce
Tree-SHA512: 3f17a91e318b9304c40c74a7a5b231149f664ae684d13e9739a05be6c05ba9720f3c3c62da6a73ace0ae8ce733f1c8410b211f9fa15694e6a8d28999ab9882d8
b298242c8d test: sqlite, add coverage for dangling to-be-reverted db txns (furszy)
fc0e747192 sqlite: guard against dangling to-be-reverted db transactions (furszy)
472d2ca981 sqlite: introduce HasActiveTxn method (furszy)
dca874e838 sqlite: add ability to interrupt statements (furszy)
fdf9f66909 test: wallet db, exercise deadlock after write failure (furszy)
Pull request description:
Discovered while was reviewing #29112, specifically https://github.com/bitcoin/bitcoin/pull/29112#pullrequestreview-1821862931.
If the db handler that initiated the database transaction is destroyed,
the ongoing transaction cannot be left dangling when the db txn fails
to abort. It must be forcefully reverted; otherwise, any subsequent
db handler executing a write operation will dump the dangling,
to-be-reverted transaction data to disk.
This not only breaks the isolation property but also results in the
improper storage of incomplete information on disk, impacting
the wallet consistency.
This PR fixes the issue by resetting the db connection, automatically
rolling back the transaction (per https://www.sqlite.org/c3ref/close.html)
when the handler object is being destroyed and the txn abortion failed.
Testing Notes
Can verify the failure by reverting the fix e5217fea and running the test.
It will fail without e5217fea and pass with it.
ACKs for top commit:
achow101:
ACK b298242c8d
ryanofsky:
Code review ACK b298242c8d. Just fix for exec result codes and comment update since last review.
Tree-SHA512: 44ba0323ab21440e79e9d7791bc1c56a8873c8bd3e8f6a85641b91576e1293011fa8032d8ae5b0580f3fb7a949356f7b9676693d7ceffa617aaad9f6569993eb
`TestNode::add_outbound_p2p_connection()` is the only place where
addconnection test-only RPC is used. here, we always pass the
appropriate v2transport option to addconnection RPC.
currently the v2transport option for addconnection RPC is optional.
so simply make the v2transport option mandatory instead.
27f260aa6e net: remove now unused global 'g_initial_block_download_completed' (furszy)
aff7d92b15 test: add coverage for peerman adaptive connections service flags (furszy)
6ed53602ac net: peer manager, dynamically adjust desirable services flag (furszy)
9f36e591c5 net: move state dependent peer services flags (furszy)
f9ac96b8d6 net: decouple state independent service flags from desirable ones (furszy)
97df4e3887 net: store best block tip time inside PeerManager (furszy)
Pull request description:
Derived from #28120 discussion.
By relocating the peer desirable services flags into the peer manager, we
allow the connections acceptance process to handle post-IBD potential
stalling scenarios.
The peer manager will be able to dynamically adjust the services flags
based on the node's proximity to the tip (back and forth). Allowing the node
to recover from the following post-IBD scenario:
Suppose the node has successfully synced the chain, but later experienced
dropped connections and remained inactive for a duration longer than the limited
peers threshold (the timeframe within which limited peers can provide blocks). In
such cases, upon reconnecting to the network, the node might only establish
connections with limited peers, filling up all available outbound slots. Resulting
in an inability to synchronize the chain (because limited peers will not provide
blocks older than the `NODE_NETWORK_LIMITED_MIN_BLOCKS` threshold).
ACKs for top commit:
achow101:
ACK 27f260aa6e
vasild:
ACK 27f260aa6e
naumenkogs:
ACK 27f260aa6e
mzumsande:
Light Code Review ACK 27f260aa6e
andrewtoth:
ACK 27f260aa6e
Tree-SHA512: 07befb9bcd0b60a4e7c45e4429c02e7b6c66244f0910f4b2ad97c9b98258b6f46c914660a717b5ed4ef4814d0dbfae6e18e6559fe9bec7d0fbc2034109200953
