fabc57e07d test: Log to debug.log in all tests (MarcoFalke)
fa4a04a5a9 test: use common setup in gui tests (MarcoFalke)
fad3d2a624 test: Create data dir in BasicTestingSetup (MarcoFalke)
Pull request description:
This makes it easier to debug a frozen test or a test that failed. To debug a failed test, remove the line `fs::remove_all(m_path_root);`.
The pull is done in three commits:
* Create a datadir for every unit test once (and only once). This requires the `SetDataDir` function to go away.
* Use the common setup in the gui unit tests. Some of those tests are testing the init sequence, so we'd have to undo some of what the testing setup did.
* Log to the debug.log in all tests
ACKs for top commit:
laanwj:
ACK fabc57e07d
Tree-SHA512: 73444210b88172669e2cd22c2703a1e30e105185d2d5f03decbdedcfd09c64ed208d3716c59c8bebb0e44214cee5c8095e3e995d049e1572ee98f1017e413665
dddd9270f8 net: Document what happens to getdata of unknonw type (MarcoFalke)
Pull request description:
Any getdata of unknown type will never be processed and blocks all future messages from a peer. This isn't obviously clear from reading the code, so document it.
Top commit has no ACKs.
Tree-SHA512: 4f8e43bbe6534242facfcfffae28b7a6aa2d228841fa2146a87d494e69f614b0da23cf7a5f3d4367358a7c1981fe2ec196a21c437ae1653f1c7e0351be22598a
099e4b9ad3 Set AA_EnableHighDpiScaling attribute early (Hennadii Stepanov)
Pull request description:
Running `bitcoin-qt` compiled against Qt 5.12.4 causes a warning:
```
hebasto@bionic-qt:~/bitcoin$ src/qt/bitcoin-qt
Attribute Qt::AA_EnableHighDpiScaling must be set before QCoreApplication is created.
```
This PR fixes this issue.
From Qt docs:
- [Qt::AA_EnableHighDpiScaling](https://doc.qt.io/qt-5/qt.html#ApplicationAttribute-enum):
> Enables high-DPI scaling in Qt on supported platforms (see also High DPI Displays). _Supported platforms are X11, Windows and Android._ Enabling makes Qt scale the main (device independent) coordinate system according to display scale factors provided by the operating system. This corresponds to setting the `QT_AUTO_SCREEN_SCALE_FACTOR` environment variable to 1. This attribute must be set before `QGuiApplication` is constructed. This value was added in Qt 5.6.
- [QCoreApplication::setAttribute()](https://doc.qt.io/qt-5/qcoreapplication.html#setAttribute)
ACKs for commit 099e4b:
MarcoFalke:
ACK 099e4b9ad3
jonasschnelli:
utACK 099e4b9ad3
fanquake:
ACK 099e4b9ad3. Did some testing on `Bionic` and `Windows 10` (using VirtualBox). I couldn't see any obvious visual difference, but given Marco's screens above, this change is obviously better. I also checked that there wasn't any sort of regression on macOS.
Tree-SHA512: 1965a427ee14ffb3871bac317685032406cf02d1fa2b2dc11c8b643bfe4ba09195674d149d1e41752f14c0d000446b35e142f3ce60d987ba97082fd7ee39a094
a2aabfb749 Use qInfo() if no error occurs (Hennadii Stepanov)
Pull request description:
[Warning and Debugging Messages](https://doc.qt.io/qt-5/debug.html#warning-and-debugging-messages):
> - `qInfo()` is used for informational messages.
> - `qWarning()` is used to report warnings and recoverable errors in your
application.
>
> If the `QT_FATAL_WARNINGS` environment variable is set, `qWarning()` exits after printing the warning message. This makes it easy to obtain a backtrace in the debugger.
[`qWarning()`](https://doc.qt.io/qt-5/qtglobal.html#qWarning):
> Calls the message handler with the warning message message... This function does nothing if `QT_NO_WARNING_OUTPUT` was defined during compilation; it exits if at the nth warning corresponding to the counter in environment variable `QT_FATAL_WARNINGS`.
This PR allows more productive debugging using the environment variable `QT_FATAL_WARNINGS`.
Examples:
- https://github.com/bitcoin/bitcoin/pull/16118#issuecomment-503184695
- https://github.com/bitcoin/bitcoin/pull/16254#issuecomment-504223404
The behavior, when option `-debug=qt` is set/unset, remains unchanged.
ACKs for commit a2aabf:
promag:
ACK a2aabfb, I also have this change locally.
Empact:
ACK a2aabfb749
laanwj:
ACK a2aabfb749
fanquake:
ACK a2aabfb749.
Tree-SHA512: b4df300c9c00a1705b0d3a10227e3deaac19a98b0a898bb60d5a88872cf450fb131eba150d9dd6c29e021566ee04b3b86b7d486bbe28bd894743c128d2309155
5224be5a33 gui: Fix open wallet menu initialization order (João Barbosa)
Pull request description:
Fixes#16230, the menu must be created before connecting to aboutToShow signal.
ACKs for commit 5224be:
hebasto:
ACK 5224be5a33, I have tested the code on Bionic with Qt 5.12.4.
ryanofsky:
utACK 5224be5a33. Looks good, fix is simple and makes perfect sense after seeing explanation in https://github.com/bitcoin/bitcoin/pull/16118#issuecomment-503166407. Without this change (and since #16118), the menu pointer passed to `connect(m_open_wallet_action->menu(), ...)` is null and connecting has no effect. With this change, the menu is constructed earlier so the connect call can work.
fanquake:
ACK 5224be5a33 Testing included in a comment above. The segfaulting with QT_FATAL_WARNINGS is unrelated to this change.
Tree-SHA512: 97b42493b37b96683058bccf39a0ee93589293d4ba8f0c60aef7f4fb9dd084cc6d5608cd5ef531cadf5e03b1f01627ef96bc2d79f784fb38cb87aa6643183d41
71d0344cf2 docs: release note wording (Karl-Johan Alm)
3d2ff37913 wallet/rpc: use static help text (Karl-Johan Alm)
53c3c1ea9e wallet/rpc/getbalances: add entry for 'mine.used' balance in results (Karl-Johan Alm)
Pull request description:
This addresses a few remaining issues pointed out in #13756:
* First commit addresses https://github.com/bitcoin/bitcoin/pull/13756#discussion_r284907468
* Second commit addresses https://github.com/bitcoin/bitcoin/pull/13756#discussion_r294868973
Ping jnewbery and achow101 as they pointed out these issues.
ACKs for commit 71d034:
jnewbery:
ACK 71d0344cf2
meshcollider:
re-utACK 71d0344cf2
Tree-SHA512: 5e28822af0574ad07dbbed21aa2fe7866bf5770b4c0a1c150ad0da8af3152bcfb7170330a7497fa500326c594740ecf63733cf58325821e2811d7b911d5783a0
Always show the same help topic regardless of wallet flags, and explain that something is not always available, rather than runtime-modifying the help output.
e61de6306f Change ismine to take a CWallet instead of CKeyStore (Andrew Chow)
7c611e2000 Move ismine to wallet module (Andrew Chow)
Pull request description:
`IsMine` isn't used outside of the wallet except for the tests. It also doesn't make sense to be outside of the wallet. This PR moves `IsMine` into the wallet module and for it to take a `CWallet` instead of `CKeyStore`. The test that used `IsMine` is also moved to the wallet tests.
This is first [prerequisites](https://github.com/bitcoin-core/bitcoin-devwiki/wiki/Wallet-Class-Structure-Changes#ismine) for the wallet structure changes.
ACKs for commit e61de6:
MarcoFalke:
re-ACK e61de6306f (only change is rebase with git auto-merge)
meshcollider:
Very light code review ACK e61de6306f
Tree-SHA512: 1cb4ad12652aef7922ab7460c6d413e8b9d1855dca78c0a286ae49d5c0765bc7996c55f262c742001d434eb9bd4215dc2cc7aae1b371ee1a82d46b32c17e6341
a49503402b Make and get the multisig redeemscript and destination in one function instead of two (Andrew Chow)
Pull request description:
`CreateMultisigRedeemscript()` is changed to `AddAndGetMultisigDestination()` so that the process of constructing the redeemScript and then getting the `CTxDestination` are done in the same function. This allows that function to see what the keys in the multisig are so that the correct address type is returned from `AddAndGetDestinationForScript()`.
This only effects the `createmultisig` and `addmultisigaddress` RPCs and does not change signing logic as #16022 does.
Alternative to #16022 and #16012Fixes#16011
ACKs for commit a49503:
Tree-SHA512: 5b0154a714deea3b2cc3a54beb420c95eeeacf4ca30c40ca80940d9d640f8b03611b0fc14c2f0710bfd8a79e8d27ad7d9ae380b4b83d52b40ab201624f2a63f0
Instead of creating a redeemScript with CreateMultisigRedeemscript and
then getting the destination with AddAndGetDestinationForScript, do
both in the same function.
CreateMultisigRedeemscript is changed to AddAndGetMultisigDestination.
It creates the redeemScript and returns it via an output parameter. Then
it calls AddAndGetDestinationForScript to add the destination to the
keystore and get the proper destination.
This allows us to inspect the public keys in the redeemScript before creating
the destination so that the correct destination is used when uncompressed
pubkeys are in the multisig.
5ebc6b0eb2 bitcoind: update -avoidpartialspends description to account for auto-enable for avoid_reuse wallets (Karl-Johan Alm)
ada258f8c8 doc: release notes for avoid_reuse (Karl-Johan Alm)
27669551da wallet: enable avoid_partial_spends by default if avoid_reuse is set (Karl-Johan Alm)
8f2e208f7c test: add test for avoidreuse feature (Karl-Johan Alm)
0bdfbd34cf wallet/rpc: add 'avoid_reuse' option to RPC commands (Karl-Johan Alm)
f904723e0d wallet/rpc: add setwalletflag RPC and MUTABLE_WALLET_FLAGS (Karl-Johan Alm)
8247a0da3a wallet: enable avoid_reuse feature (Karl-Johan Alm)
eec15662fa wallet: avoid reuse flags (Karl-Johan Alm)
58928098c2 wallet: make IsWalletFlagSet() const (Karl-Johan Alm)
129a5bafd9 wallet: rename g_known_wallet_flags constant to KNOWN_WALLET_FLAGS (Karl-Johan Alm)
Pull request description:
Add a new wallet flag called `avoid_reuse` which, when enabled, will keep track of when a specific destination has been spent from, and will actively "blacklist" any new UTXOs which send to an already-spent-from destination.
This improves privacy, as a payer could otherwise begin tracking a payee's wallet by regularly peppering a known UTXO with dust outputs, which would then be scooped up and used in payments by the payee, allowing the payer to map out (1) the inputs owned by the payee and (2) the destinations to which the payee is making payments.
This replaces #10386 and together with the (now merged) #12257 it addresses #10065 in full. The concerns raised in https://github.com/bitcoin/bitcoin/pull/10386#issuecomment-302361381 are also addressed due to #12257.
~~Note: this builds on top of #15780.~~ (merged)
ACKs for commit 5ebc6b:
jnewbery:
ACK 5ebc6b0eb
laanwj:
Concept and code-review ACK 5ebc6b0eb2
meshcollider:
Code review ACK 5ebc6b0eb2
achow101:
ACK 5ebc6b0eb2 modulo above nits
Tree-SHA512: fdef45826af544cbbb45634ac367852cc467ec87081d86d08b53ca849e588617e9a0a255b7e7bb28692d15332de58d6c3d274ac003355220e4213d7d9070742e
0959d37e3e Don't use global (external) symbols for symbols that are used in only one translation unit (practicalswift)
Pull request description:
Don't use global (external) symbols for symbols that are used in only one translation unit.
Before:
```
$ for SYMBOL in $(nm src/bitcoind | grep -E ' [BD] ' | c++filt | cut -f3- -d' ' | grep -v @ | grep -v : | sort | grep '[a-z]' | sort -u | grep -vE '(^_|typeinfo|vtable)'); do
REFERENCES=$(git grep -lE "([^a-zA-Z]|^)${SYMBOL}([^a-zA-Z]|\$)" -- "*.cpp" "*.h")
N_REFERENCES=$(wc -l <<< "${REFERENCES}")
if [[ ${N_REFERENCES} > 1 ]]; then
continue
fi
echo "Global symbol ${SYMBOL} is used in only one translation unit: ${REFERENCES}"
done
Global symbol g_chainstate is used in only one translation unit: src/validation.cpp
Global symbol g_ui_signals is used in only one translation unit: src/ui_interface.cpp
Global symbol instance_of_cmaincleanup is used in only one translation unit: src/validation.cpp
Global symbol instance_of_cnetcleanup is used in only one translation unit: src/net.cpp
Global symbol instance_of_cnetprocessingcleanup is used in only one translation unit: src/net_processing.cpp
Global symbol pindexBestForkBase is used in only one translation unit: src/validation.cpp
Global symbol pindexBestForkTip is used in only one translation unit: src/validation.cpp
$
```
After:
```
$ for SYMBOL in $(nm src/bitcoind | grep -E ' [BD] ' | c++filt | cut -f3- -d' ' | grep -v @ | grep -v : | sort | grep '[a-z]' | sort -u | grep -vE '(^_|typeinfo|vtable)'); do
REFERENCES=$(git grep -lE "([^a-zA-Z]|^)${SYMBOL}([^a-zA-Z]|\$)" -- "*.cpp" "*.h")
N_REFERENCES=$(wc -l <<< "${REFERENCES}")
if [[ ${N_REFERENCES} > 1 ]]; then
continue
fi
echo "Global symbol ${SYMBOL} is used in only one translation unit: ${REFERENCES}"
done
$
```
♻️ Think about future generations: save the global namespace from unnecessary pollution! ♻️
ACKs for commit 0959d3:
Empact:
ACK 0959d37e3e
MarcoFalke:
ACK 0959d37e3e
hebasto:
ACK 0959d37e3e
promag:
ACK 0959d37.
Tree-SHA512: 722f66bb50450f19b57e8a8fbe949f30cd651eb8564e5787cbb772a539bf3a288c048dc49e655fd73ece6a46f6dafade515ec4004729bf2b3ab83117b7c5d153
faa2a47cd7 logging: Add threadsafety comments (MarcoFalke)
0b282f9b00 Log early messages with -printtoconsole (Anthony Towns)
412987430c Replace OpenDebugLog() with StartLogging() (Anthony Towns)
Pull request description:
Early log messages are dropped on the floor and they'd never make it to the console or debug log. This can be tested by running the test included in this pull request without re-compiling the `bitcoind`.
Fix that by buffering early messages and flushing them as soon as all logging options have been initialized and logging has been started.
This pull request is identical to "Log early messages with -printtoconsole" (#13088) by **ajtowns**, with the following changes:
* Rebased
* Added docstrings for `m_buffering` and `StartLogging`
* Switch `CCriticalSection` (aka `RecursiveMutex`) to just `Mutex` in the last commit
* Added tests
Fixes#16098Fixes#13157Closes#13088
ACKs for commit faa2a4:
ajtowns:
utACK faa2a47cd7
hebasto:
ACK faa2a47cd7
kristapsk:
ACK faa2a47cd7 (ran added functional test before / after recompiling, didn't do additional testing)
Tree-SHA512: 685e2882642fe2a43ce171d42862582dadb840d03cda8236a994322c389ca2a1f3f431b179b2726c155c61793543bb340c568a5455d97f8b83bc7d307a85d387
8a2656702b torcontrol: Use the default/standard network port for Tor hidden services, even if the internal port is set differently (Luke Dashjr)
Pull request description:
Currently, the hidden service is published on the same port as the public listening port.
But if a non-standard port is configured, this can be used to guess (pretty reliably) that the public IP and the hidden service are the same node.
ACKs for top commit:
practicalswift:
utACK 8a2656702b
naumenkogs:
utACK 8a26567
laanwj:
utACK 8a2656702b
Tree-SHA512: 737c8da4f7c3f0bb22a338647d357987f5808156e3f38864168d0d8c2e2b171160812f7da4de11eef602902b304e357d76052950b72d7b3b83535b0fdd05fadc
86b47fa741 speed up Unserialize_impl for prevector (Akio Nakamura)
Pull request description:
The unserializer for prevector uses `resize()` for reserve the area, but it's prefer to use `reserve()` because `resize()` have overhead to call its constructor many times.
However, `reserve()` does not change the value of `_size` (a private member of prevector).
This PR make the logic of read from stream to callback function, and prevector handles initilizing new values with that call-back and ajust the value of `_size`.
The changes are as follows:
1. prevector.h
Add a public member function named 'append'.
This function has 2 params, number of elemenst to append and call-back function that initilizing new appended values.
2. serialize.h
In the following two function:
- `Unserialize_impl(Stream& is, prevector<N, T>& v, const unsigned char&)`
- `Unserialize_impl(Stream& is, prevector<N, T>& v, const V&)`
Make a callback function from each original logic of reading values from stream, and call prevector's `append()`.
3. test/prevector_tests.cpp
Add a test for `append()`.
## A benchmark result is following:
[Machine]
MacBook Pro (macOS 10.13.3/i7 2.2GHz/mem 16GB/SSD)
[result]
DeserializeAndCheckBlockTest => 22% faster
DeserializeBlockTest => 29% faster
[before PR]
# Benchmark, evals, iterations, total, min, max, median
DeserializeAndCheckBlockTest, 60, 160, 94.4901, 0.0094644, 0.0104715, 0.0098339
DeserializeBlockTest, 60, 130, 65.0964, 0.00800362, 0.00895134, 0.00824187
[After PR]
# Benchmark, evals, iterations, total, min, max, median
DeserializeAndCheckBlockTest, 60, 160, 77.1597, 0.00767013, 0.00858959, 0.00805757
DeserializeBlockTest, 60, 130, 49.9443, 0.00613926, 0.00691187, 0.00635527
ACKs for top commit:
laanwj:
utACK 86b47fa741
Tree-SHA512: 62ea121ccd45a306fefc67485a1b03a853435af762607dae2426a87b15a3033d802c8556e1923727ddd1023a1837d0e5f6720c2c77b38196907e750e15fbb902
8053e5cdad Remove -mempoolreplacement to prevent needless block prop slowness. (Matt Corallo)
Pull request description:
At this point there is no reasonable excuse to disable opt-in RBF,
and, unlike when this option was added, there are now significant
issues created when disabling it (in the form of compact block
reconstruction failures). Further, it breaks a lot of modern wallet
behavior.
This removes an option that is:
* (a) only useful when a large portion of (other) miners enforce it as well
* (b) is detrimental to everyone (income for miners, RBF notifications for others) who uses it individually otherwise
* (c) is effectively unused
* (d) is often confused with disabling RBF (rather than just remaining stubbornly unaware of it while the rest of the network lets it through)
ACKs for commit 8053e5:
practicalswift:
utACK 8053e5cdad
promag:
Deprecation would save from unlikely rantings, still ACK 8053e5c.
jtimon:
utACK 8053e5cdad
ajtowns:
ACK 8053e5cdad -- quick code review, checked tests work
MarcoFalke:
ACK 8053e5cdad
Tree-SHA512: 01aee8905b2487fc38a3a86649d422d2d2345bc60f878889ebda4b8680783e1f1a97c2000c27ef086719501be2abc2911b2039a259a5e5c04f3b24ff02b0427e
fa499b5f02 rpc: bugfix: Properly use iswitness in converttopsbt (MarcoFalke)
fa5c5cd141 rpc: Switch touched RPCs to IsValidNumArgs (MarcoFalke)
Pull request description:
When a serialized transaction has inputs, there is no risk in only trying to deserialize it with witness allowed. (This is how all transactions from p2p are deserialized.) In fact, it would avoid a common issue where a transaction with inputs can be deserialized in two ways:
* Fixes#12989
* Fixes#15872
* Fixes#15701
* Fixes#13738
* ...
When a serialized transaction has no inputs, there is no risk in only trying to deserialze it with witness disallowed. (A transaction without inputs can't have corresponding witness data)
ACKs for commit fa499b:
meshcollider:
utACK fa499b5f02
ryanofsky:
utACK fa499b5f02. Changes since last review: consolidating commits and making iswitness documentation the same across methods.
PastaPastaPasta:
utACK fa499b5f02
Tree-SHA512: a64423a3131f3f0222a40da557c8b590c9ff01b45bcd40796f77a1a64ae74c6680a6be9d01ece95c492dfbcc7e2810409d2c2b336c2894af00bb213972fc85c6
c59e3a3261 getrawtransaction: inform about blockhash argument when lookup fails (darosior)
Pull request description:
Just 4 words added on `getrawtransaction` lookup error to fix#16142
ACKs for commit c59e3a:
Tree-SHA512: 2219099c1240667527a9b1498a58818b5ff1c2ef366c498d2bb57963e828b3c87fa3e6b94be7e6463bd289ceabc13f9c9b1082134641594ba335ac400e6d63aa
fa8f195195 Replace remaining fprintf with tfm::format manually (MarcoFalke)
fac03ec43a scripted-diff: Replace fprintf with tfm::format (MarcoFalke)
fa72a64b90 tinyformat: Add doc to Bitcoin Core specific strprintf (MarcoFalke)
Pull request description:
This should be a refactor except in the cases where we use the wrong format specifier [1], in which case this patch is a bug fix.
[1] : e.g. depends: Add libevent compatibility patch for windows #8730
ACKs for commit fa8f19:
promag:
ACK fa8f195195. Ideally this should be rebased before merge.
practicalswift:
utACK fa8f195195
Empact:
ACK fa8f195195
laanwj:
code review and lightly tested ACK fa8f195195
jonatack:
ACK fa8f195195 from light code review, building, and running linter/unit tests/extended functional tests.
Tree-SHA512: 65f648b0bc383e3266a5bdb4ad8c8a1908a719635d49e1cd321b91254be24dbc7e22290370178e29b98ddcb3fec0889de9cbae273c7140abc9793d849534a743
b748bf6f50 Fix spelling errors identified by codespell 1.15.0 (Ben Woosley)
Pull request description:
Note all changes are to comments / documentation.
After this commit, the only remaining output is:
```
$ test/lint/lint-spelling.sh
src/test/base32_tests.cpp:14: fo ==> of, for
src/test/base64_tests.cpp:14: fo ==> of, for
^ Warning: codespell identified likely spelling errors. Any false positives? Add them to the list of ignored words in test/lint/lint-spelling.ignore-words.txt
```
Note:
* I ignore several valid alternative spellings ~, but changed homogenous
to homogeneous as the latter is a more specific term according to the
Google dictionary definitions I found~
* homogenous is present in tinyformat, hence should be addressed upstream
* process' is correct only if there are plural processes
ACKs for commit b748bf:
practicalswift:
ACK b748bf6f50
fanquake:
ACK b748bf6f50
Tree-SHA512: 9add7044643ce015e0a44d8b27a3f300d72c485ffff550fb6491a17f14528085289ec5caddfe02f291ea9b2cded38a0dd3079652a054e2d7fe2ff4f7b53db5d7
f402012cc fixup: Fix prunning test (João Barbosa)
97f517dd8 Fix RPC/pruneblockchain returned prune height (Jonas Schnelli)
Pull request description:
The help of `pruneblockchain` tells us that the return value is `Height of the last block pruned.`,... but the implementation naively returns the provided input `height` and therefore not respecting that pruning can't be done on all possible blockheight due to the fact that we only prune complete blockfiles (which combine multiple blocks).
This fixes the return value to actually return the correct prune height.
ACKs for commit f40201:
MarcoFalke:
ACK f402012ccf
Tree-SHA512: 88c910030ffb83196663e5ebebc29d036fcdbbb2ab266e4538991867924a61bacd8361c1fbf294a0ea7e02347ae183d792f10a10b8f6187e8a4c4c6e4124d7e6
75485ef09 gui: Enable open wallet menu on setWalletController (João Barbosa)
Pull request description:
`BitcoinApplication::initializeResult` and `BitcoinGUI::setWalletController` are only called after the startup rescan is completed. While the rescan is in progress the window menus are already available.
This PR makes the Open Wallet menu disabled until `BitcoinGUI::setWalletController` is called.
Fixes#16087
ACKs for commit 75485e:
jonasschnelli:
utACK 75485ef096
ryanofsky:
utACK 75485ef096. It's a simple, sensible fix.
Tree-SHA512: 9395ceed54bbceb6cbf1cd443f783d07a6ebb8fc5515b63c6e1b8b19b216b08d1cba7eaf872814d7c426ab7192f3b416ba0d57fc84f3bcbfebf01ce153794201
308b76732f Fix bug around transaction requests (Suhas Daftuar)
f635a3ba11 Expire old entries from the in-flight tx map (Suhas Daftuar)
e32e08407e Remove NOTFOUND transactions from in-flight data structures (Suhas Daftuar)
23163b7593 Add an explicit memory bound to m_tx_process_time (Suhas Daftuar)
218697b645 Improve NOTFOUND comment (Suhas Daftuar)
Pull request description:
#14897 introduced several bugs that could lead to a node no longer requesting transactions from one or more of its peers. Credit to ajtowns for originally reporting many of these bugs along with an originally proposed fix in #15776.
This PR does a few things:
- Fix a bug in NOTFOUND processing, where the in-flight map for a peer was keeping transactions it shouldn't
- Eliminate the possibility of a memory attack on the CNodeState `m_tx_process_time` data structure by explicitly bounding its size
- Remove entries from a peer's in-flight map after 10 minutes, so that we should always eventually resume transaction requests even if there are other bugs like the NOTFOUND one
- Fix a bug relating to the coordination of request times when multiple peers announce the same transaction
The expiry mechanism added here is something we'll likely want to remove in the future, but is belt-and-suspenders for now to try to ensure we don't have other bugs that could lead to transaction relay failing due to some unforeseen conditions.
ACKs for commit 308b76:
ajtowns:
utACK 308b76732f
morcos:
light ACK 308b767
laanwj:
Code review ACK 308b76732f
jonatack:
Light ACK 308b76732f.
jamesob:
ACK 308b76732f
MarcoFalke:
ACK 308b76732f (Tested two of the three bugs this pull fixes, see comment above)
jamesob:
Concept ACK 308b76732f
MarcoFalke:
ACK 308b76732f
Tree-SHA512: 8865dca5294447859d95655e8699085643db60c22f0719e76e961651a1398251bc932494b68932e33f68d4f6084579ab3bed7d0e7dd4ac6c362590eaf9414eda
After this commit, the only remaining output is:
$ test/lint/lint-spelling.sh
src/test/base32_tests.cpp:14: fo ==> of, for
src/test/base64_tests.cpp:14: fo ==> of, for
^ Warning: codespell identified likely spelling errors. Any false positives? Add them to the list of ignored words in test/lint/lint-spelling.ignore-words.txt
Note:
* I ignore several valid alternative spellings
* homogenous is present in tinyformat, hence should be addressed upstream
* process' is correct only if there are plural processes
At this point there is no reasonable excuse to disable opt-in RBF,
and, unlike when this option was added, there are now significant
issues created when disabling it (in the form of compact block
reconstruction failures). Further, it breaks a lot of modern wallet
behavior.
