4e29de719e private broadcast: add release note for limited cap (Gregory Sanders)
cbf8c107c1 Release cs_main between individual private tx re-attempts (Greg Sanders)
5aea3d0373 private broadcast: limit outstanding txs to count of 10,000 (Gregory Sanders)
Pull request description:
Add a belt-and-suspenders feature, limit the amount of memory and cpu possible when unlucky or simply misconfigured. The worst case limit is roughly 400kB * 10,000 = 4GB, regardless of usage pattern.
Before this change, sheer volume of broadcasts, mismatches in standardness rules, or simply fee mismatches may result in unbounded growth of memory usage. As the feature may be expanded in the future, explicit bounds helps reasoning going forward.
ACKs for top commit:
frankomosh:
tACK 4e29de719e. Ran private_broadcast_tests and p2p_private_broadcast_cap.py. Great to have an explicit bound as the belt-and-suspenders against unbounded queue growth.
vasild:
ACK 4e29de719e
andrewtoth:
ACK 4e29de719e
stickies-v:
ACK 4e29de719e
Tree-SHA512: 18161755f37d07cca185a09e782dbe2fd0025b8befd4f6660e988865cc3a9b705d41769b816161e8142fe6ce31a56e0288bd78efc25135cedfc47fc855011799
55e3a57f22 qa: Avoid UTXO reuse between test functions (Hodlinator)
9c5dd2926a p2p: Ignore CMPCTBLOCK from peer that hasn't sent SENDCMPCT (David Gumberg)
bf9884f4e5 p2p: make blocksonly nodes ignore CMPCTBLOCK messages (David Gumberg)
92cea63c71 test: (Un)solicited invalid cb -> get disconnected. (David Gumberg)
e845e26344 test: p2p: Nodes ignore unsolicited CMPCTBLOCK's (David Gumberg)
8313591715 p2p: Drop unsolicited CMPCTBLOCK from non-HB peer (David Gumberg)
44f377a71f refactor: test: Static assert_highbandwidth_states (David Gumberg)
25457a3272 test: Tighten getblocktxn checks in parallel cb reconstruction test. (David Gumberg)
51dd90fb50 refactor: Merge announce_cmpct_block() defs into one (Hodlinator)
Pull request description:
Processing unsolicited `CMPCTBLOCK`'s from a peer that has not been marked high bandwidth is not well-specified behavior in BIP-0152, in fact the BIP seems to imply that it is not permitted:
> "[...] method is not useful for compact blocks because `cmpctblock` blocks can be sent unsolicitedly in high-bandwidth mode"
See https://github.com/bitcoin/bips/blob/master/bip-0152.mediawiki#separate-version-for-segregated-witness
This PR disables processing of CMPCTBLOCK messages in three cases:
$1$. When the block is unsolicited and from a non-HB peer.
$2$. When this node is running in `-blocksonly` mode.
$3$. When the peer has not advertised `CMPCTBLOCK` support with a `SENDCMPCT` message.
Not processing unsolicited blocks slightly raises the cost of discovering a peer's mempool via `CMPCTBLOCK` as described in #28272. As pointed out there, getting an HB slot is relatively easy, so this does not prevent an attacker from doing this, it just slightly raises the bar.
Probably more important is not processing `CMPCTBLOCK` messages as a `-blocksonly` node. A blocksonly node has a lot less surface area for leaking its mempool since it does no transaction relay, and leaking a blocksonly node's mempool is pretty dangerous since it is very likely to be the origin for all of the transactions in its mempool.
ACKs for top commit:
achow101:
ACK 55e3a57f22
w0xlt:
reACK 55e3a57f22
hodlinator:
re-ACK 55e3a57f22
polespinasa:
lgtm ACK 55e3a57f22
Tree-SHA512: 118bea55adca01dbd6467ba5ae3adf420d960794a6a2c40dd30fcc7d79aa944e01af0f6dd6bd6ff6d33dc9155171f6f4f497cbd7e8eb6d3c4c89e12740b51c05
c1313b199f init: wake genesis wait after ImportBlocks() returns (ismaelsadeeq)
Pull request description:
During startup with `-reindex`, the block index and chainstate are wiped, and
the normal `LoadGenesisBlock()` startup path is skipped while block files are
being indexed. The init thread can then wait for genesis to be processed via the
tip-block condition variable.
If shutdown is requested after the init thread enters that wait but before the
import thread activates genesis, the reindex scan can return early without
calling the later `LoadGenesisBlock()` fallback or `ActivateBestChains()`.
Because no block tip is connected, the block-tip notification never fires. The
wait predicate would accept `ShutdownRequested(node)`, but it is not evaluated
again unless the condition variable is notified.
Notify the tip-block condition variable after `ImportBlocks()` returns so the
genesis wait can observe shutdown and exit cleanly.
Described in detail https://github.com/bitcoin/bitcoin/pull/35621#issuecomment-4876059120
The current test covers the path, but won't fail deterministically. You can use L0rinc suggested patch for a deterministic patch that demonstrate the deadlock on master https://github.com/bitcoin/bitcoin/pull/35652#pullrequestreview-4627988553
ACKs for top commit:
maflcko:
review ACK c1313b199f🦁
mzumsande:
Code Review ACK [c1313b1](c1313b199f)
sedited:
tACK c1313b199f
Tree-SHA512: b17faaa2042882f2acebf688f4a28fd6bb11cc5d7f978991fe2bcba7a9df6ed82a0a2811f2e71a929189570a7926f148afdbeef5c1ca57a081e3bb914c2f6fe2
If shutdown interrupts ImportBlocks() before genesis activation,
no blockTip notification is sent. The wait predicate allows
shutdown to occur, but the condition variable is never
notified, so init can remain stuck waiting for genesis activation.
Notify the tip condition variable after ImportBlocks() returns so
interrupted imports wake the wait and let it observe the shutdown request.
Add test coverage for interrupting startup after reindex block files import
begins, which exercises the path where ImportBlocks() can return before genesis
activation.
9a8ef9b0a3 test: SOCKS5 proxy: expect that connection may be reset during handshake (Vasil Dimov)
eb3208364a test: SOCKS5 proxy: expect that connection may be reset when forwarding (Vasil Dimov)
Pull request description:
The `forward_sockets()` function used by the SOCKS5 proxy forwards data between two connected sockets. It might happen that one of those sockets gets closed/reset abruptly, without sending EOF first. This is to be expected if e.g. `bitcoind` is shutdown and shouldn't result in noisy harmless messages like:
```
2026-06-03T13:23:56.966859Z TestFramework.socks5 (ERROR): socks5 request handling failed (running True)
Traceback (most recent call last):
File ".../socks5.py", line 199, in handle
forward_sockets(self.conn, conn_to, self.wakeup_socket_pair[1], self.serv)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File ".../socks5.py", line 76, in forward_sockets
data = s.recv(4096)
ConnectionResetError: [Errno 104] Connection reset by peer
```
Instead turn this into a debug log message with a nice prefix containing enough information to identify the two forwarded sockets.
---
Also expect that the connection might be closed during the SOCKS5 handshake and only log a debug message if that happens.
ACKs for top commit:
optout21:
crACK 9a8ef9b0a3
danielabrozzoni:
reACK 9a8ef9b0a3
sedited:
ACK 9a8ef9b0a3
Tree-SHA512: 24e25a30529eda3536ebf472f63a93fd80fff46273054a7075490c88737f8870c0141b2bc99d9ef39e6b4f592af2801350fdfbc71927f573738b4a14f5fd7ce0
a15bdc0598 doc: update offline-signing-tutorial to use exportwatchonlywallet rpc (Pol Espinasa)
a388076401 test: Test for exportwatchonlywallet (Ava Chow)
d053e3e5c8 wallet, rpc: Add exportwatchonlywallet RPC (Ava Chow)
444878efef wallet: Add CWallet::ExportWatchOnly (Ava Chow)
f9273f01db wallet: Move listdescriptors retrieving from RPC to CWallet (Ava Chow)
a1c83789a7 wallet: Write new descriptor's cache in AddWalletDescriptor (Ava Chow)
1e996640e6 wallet: Use Descriptor::CanSelfExpand() in CanGetAddresses() (Ava Chow)
d2ee9227da descriptor: Add CanSelfExpand() (Ava Chow)
Pull request description:
Currently, if a user wants to use an airgapped setup, they need to manually create the watchonly wallet that will live on the online node by importing the public descriptors. This PR introduces `exportwatchonlywallet` which will create a wallet file with the public descriptors to avoid exposing the specific internals to the user. Additionally, this RPC will copy any existing labels, transactions, and wallet flags. This ensures that the exported watchonly wallet is almost entirely a copy of the original wallet but without private keys.
ACKs for top commit:
polespinasa:
lgtm ACK a15bdc0598
Sjors:
re-utACK a15bdc0598
pablomartin4btc:
re-ACK [a15bdc0](a15bdc0598)
w0xlt:
lgtm reACK a15bdc0598
Tree-SHA512: cfc59415ad9aa13d1445cf2a85db1c051215496b6edcf5a8db463499b2b51b92ee7bf840b709035dff7635f9d0c533423bceb58c851f220500e1ea254d12f3b8
Introduce a new simple `NodeSigner` wallet replacement that delegates
signing to a specified node via the `signrawtransactionwithkey` RPC,
and plug in that new class to the functional test feature_taproot.py,
so that it can be ran without the Bitcoin Core wallet compiled.
This can be useful for spending funds that are received at the created
address. Preparatory for the next commit, which modifies the taproot
functional test to work without the Bitcoin Core wallet compiled in.
Add a belt-and-suspenders feature, limit the amount of
memory and cpu possible when unlucky or simply misconfigured.
The worst case limit is roughly 400kB * 10,000 = 4GB, regardless
of usage pattern.
Before this change, sheer volume of broadcasts, mismatches in
standardness rules, or simply fee mismatches may result in unbounded
growth of memory usage. As the feature may be expanded in
the future, explicit bounds helps reasoning going forward.
3765b428d1 logging: More fully remove libevent log category (Ryan Ofsky)
Pull request description:
Libevent log category was partially removed in 39e9099da5, and this commit extends that with the following changes:
- Stops showing libevent in the list of supported log categories in `bitcoind -help` and `bitcoin-cli help logging` output.
- Stops returning `"libevent": false` in `logging` RPC output.
It's not good to treat libevent as a supported log category when it can't be enabled and trying to enable it results in warnings.
There's also no need to define an unused LIBEVENT constant value and keep more complicated logic for dealing with deprecated log categories, so this change also simplifies code internally.
ACKs for top commit:
l0rinc:
code review ACK 3765b428d1
pinheadmz:
ACK 3765b428d1
sedited:
ACK 3765b428d1
Tree-SHA512: 09e9514f905bb0a79d870689af491886baaa31fa19f2ad6aef4283fa20c2fa6ce8d384178139227aeeabffabff6e83d254114daaeefbbfe2c6172b9da8871298
8791c4764c test: use ExtendedPrivateKey in wallet_taproot.py (rkrux)
89ceafafb9 test: use ExtendedPrivateKey in wallet_listdescriptors.py (rkrux)
bbfffcab58 test: use ExtendedPrivateKey in wallet_send.py (rkrux)
2ab6e590f7 test: use ExtendedPrivateKey in wallet_keypool.py (rkrux)
9e20118720 test: use ExtendedPrivateKey in wallet_fundrawtransaction.py (rkrux)
06af0cddbb test: use ExtendedPrivateKey in wallet_descriptor.py (rkrux)
4100fac20e test: use ExtendedPrivateKey in wallet_createwallet.py (rkrux)
ff3f6def9a test: use ExtendedPrivateKey in wallet_bumpfee.py (rkrux)
003f2a01f6 test: use ExtendedPrivateKey in feature_notifications.py (rkrux)
f988e6d6e6 test: use ExtendedPrivateKey in wallet_importdescriptors.py (rkrux)
d2a03d50ac test: add extendedkey.py unit tests by using BIP32 test vectors (rkrux)
afdb378082 test: introduce ExtendedPrivateKey and ExtendedPublicKey classes (rkrux)
4dbaa7cc65 test: generalise byte_to_base58 utility function to allow more version types (rkrux)
Pull request description:
Many a times there has been a need to come up with dynamic xprvs and xpubs
in the functional tests, but the lack of code that creates them dynamically has
led to the presence of several hardcoded keys in the testing framework. This
is not developer friendly and not self-documenting, clutters the testing code,
and makes it difficult to update the tests in the future.
This PR introduces two utility classes ExtendedPrivateKey and
ExtendedPublicKey that allows the developer to create them on the fly
to be used in the tests. I have intentionally not introduced any library for this
purpose and have reused the existing libraries and functions in the framework.
The implementation is supposed to provide basic functionality for creating
xprv randomly or from a fixed seed, creating corresponding xpub, and
deriving child xprvs and xpubs at custom derivation paths.
I've updated many tests to show how these can be used, there are more
tests as well that can be updated in the future to completely remove such
non-deterministic hardcoded keys.
ACKs for top commit:
achow101:
ACK 8791c4764c
w0xlt:
ACK 8791c4764c
Tree-SHA512: f8ec4e09eaa6cc44b0f1c9a91337e570b12fb882c258be89b470de1a8cecf9d2fd40d9f02ee739dcbf639462ea7710aa145a3726f0f537f5a1f1e7772e5b019d
9e6546c517 test: raise reindex mining RPC timeout (Lőrinc)
Pull request description:
**Problem:** I often hit a timeout in `feature_reindex.py` when running functional tests locally in parallel in debug mode (especially on battery or in power-saving mode).
**Fix:** Increase the test-local RPC timeout for the reindex mining setup.
ACKs for top commit:
mercie-ux:
ACK 9e6546c517
sedited:
ACK 9e6546c517
Tree-SHA512: d3541dd6752f943921a030393b00e684b3b5d00b93aa0b2b1f85c017698bdfe2216d7f5441bb7dd5f153453385e4df6f9cc8055589e57bca83fa48f0b7de4252
0cdd817a82 add release note (Pol Espinasa)
517d37ce3e test: tests wallet migration with load_wallet disabled (Pol Espinasa)
b98dd63da7 rpc: Add load_wallet argument to migratewallet RPC (Pol Espinasa)
4acd063ba6 wallet: make loading the wallet after migrating optional (Pol Espinasa)
97d08d62ba refactor: store wallet names to MigrationResult (Pol Espinasa)
Pull request description:
This PR is motivated by this [Stack Exchange question](https://bitcoin.stackexchange.com/questions/130713/bitcoin-core-quickest-method-legacy-descriptor-wallet-migration).
Long story short, someone who has a node pruned before his legacy wallet birthday, is unable to migrate the wallet as it is not possible to load it.
Loading is not necessary for migration, and migrating without wanting to use the wallet in that node is a valid use-case.
This PR adds a new RPC argument to `migratewallet` that allow the user disabling the wallet loading.
Second commits adds tests for it.
Follow-up: Add an option to the GUI to not load the wallet after migrating.
ACKs for top commit:
achow101:
ACK 0cdd817a82
w0xlt:
ACK 0cdd817a82
pablomartin4btc:
ACK 0cdd817a82
Tree-SHA512: 8389599e63603b1a532e1bfba0b6c652653386c001f5a881bd49843302b74ff4dbaa4131b5b377c24f483d42e0e70a92b96f760244e3c2e2b44ce08cd04ca1e0
Libevent log category was partially removed in 39e9099da5, and this
commit extends that with the following changes:
- Stops showing libevent in the list of supported log categories in
`bitcoind -help` and `bitcoin-cli help logging` output.
- Stops returning `"libevent": false` in `logging` RPC output.
It's not good to treat libevent as a supported log category when it
can't be enabled and trying to enable it results in warnings.
There's also no need to define an unused LIBEVENT constant value and
keep more complicated logic for dealing with deprecated log categories,
so this change also simplifies code internally.
Co-authored-by: David Gumberg <davidzgumberg@gmail.com>
Co-authored-by: l0rinc <pap.lorinc@gmail.com>
abc33ff043 test: announce field must be 0 or 1 in sendcmpct (brunoerg)
2d0dce0af5 net_processing: fix BIP152 first integer interpretation (brunoerg)
Pull request description:
Fixes#35542
According to the BIP152, the first integer in `sendcmpct` message shall be interpreted as a boolean (and MUST have a value of either 1 or 0). We currently correctly interpret it as boolean, however, we accept any value >=1 and treat it as `true`, deviating from the specification. This PR fixes it.
ACKs for top commit:
edilmedeiros:
utACK abc33ff043
davidgumberg:
crACK abc33ff043 Seems reasonable to comply with BIP152 strictly, test looks good as well.
Sjors:
ACK abc33ff043
jonatack:
re-ACK abc33ff043
achow101:
ACK abc33ff043
w0xlt:
ACK abc33ff043
Tree-SHA512: 77fed86d4de81f7c35ff002b6e1b2a90882ea55f159075da4d34a619d1075f625fca34f929cdd981f1b2eb06f76b64f42cda46502dcd1b4a634c690b0882ec7c
b847626562 test: refresh MiniWallet after node restart (Sjors Provoost)
f4e643cb15 test: merge mining options in package feerate check (Sjors Provoost)
280ce6a0ae miner: ensure block_max_weight is flattened before limit checks (Sjors Provoost)
65bd3164fb mining: clarify test_block_validity comment (Sjors Provoost)
978e7216e6 test: use shared default_ipc_timeout (Sjors Provoost)
Pull request description:
This implement the suggested followups from #33966. Each commit links to the original comment.
The most important change is the extra asserts added in `miner: ensure block_max_weight is flattened before limit checks`.
ACKs for top commit:
achow101:
ACK b847626562
enirox001:
tACK b847626562
sedited:
ACK b847626562
w0xlt:
ACK b847626562
Tree-SHA512: 47678eaed604228269bd892ccf8ff58804745bbc7675b4a93528da9a9292a2eb1e0562cdb8341edac77178563420885b48282bb9e5c2b997b28f2fc64ceeff3d
2fe34808fa wallet: reject sendtoaddress and sendmany for external signers (Sjors Provoost)
bd5a32f7db doc: add taproot descriptor to getdescriptors example (woltx)
7131c82937 doc: clarify which commands receive --chain, --fingerprint and --stdin (woltx)
4fdd4d8d29 doc: replace stale signtransaction wording with current signtx flow (woltx)
fab92257fe doc, rpc: document enumerate model field and fingerprint deduplication (woltx)
Pull request description:
This PR aligns the external signer documentation with current behavior, and makes one previously implicit behavior explicit.
Per review feedback, each commit fixes a limited set of issues:
* **doc, rpc: document enumerate model field and fingerprint deduplication** — the `enumerate` response uses the optional `model` field, which Bitcoin Core maps to the `name` field of the `enumeratesigners` RPC result. Duplicate fingerprints are skipped, and wallet operations require exactly one connected signer.
* **doc: replace stale signtransaction wording with current signtx flow** — spending from an external signer wallet uses `send`/`sendall` (and `bumpfee` for fee-bumping), which invoke `<cmd> --stdin` and pass the `signtx` subcommand and PSBT over stdin.
* **doc: clarify which commands receive --chain, --fingerprint and --stdin** — mark `--chain` and `--fingerprint` as required except for `enumerate`, keep `--stdin` required for protocol flexibility, and match the order and form of the actual invocations in the usage examples.
* **doc: add taproot descriptor to getdescriptors example** — show the BIP86 `tr()` descriptor alongside the other address types.
* **wallet: reject sendtoaddress and sendmany for external signers** — return a specific error instead of the misleading "Private keys are disabled for this wallet", with functional test coverage. Cherry-picked from #33112 (thanks Sjors).
How the documentation went stale:
* The `enumerate` example has shown a `name` field since external signer support landed in #16546, but the implementation has always read `model`.
* `sendtoaddress`/`sendmany` external signer support was effectively precluded by #21201, which was merged a few days before #16546, so the interaction was missed in review and the documented `signtransaction` flow never existed in this form.
* Fingerprint deduplication was added in #35251.
* The documentation was last updated in #33765.
ACKs for top commit:
Sjors:
ACK 2fe34808fa
optout21:
ACK 2fe34808fa
naiyoma:
ACK 2fe34808fa
Tree-SHA512: 86859d2f81ac337f3b4b6578c6ee0151ffb76b8374dfa58e28e00ce4eb69dc200cd6bd2d0a99f73d0475c3824d6ac1cb9e2542b119ca124dd835132dc95cd023
This avoids hardcoding xprvs and xpubs in the test.
Also use key's parent fingerprint and derivation path while creating descriptors
so that all of it is self-documenting and not dependent on hardcoded values,
making them easy to read and update.
Also use descsum_create function more to avoid hardcoding the descriptor
checksum.
d6359937bf validation: check invariants when inserting into m_blocks_unlinked (stratospher)
0852925bd8 test/doc: remove misleading comment and improve tests (stratospher)
ca4a380281 test: add coverage for UB caused by FindMostWorkChain (stratospher)
c787b3b99b validation: avoid duplicates in m_blocks_unlinked (stratospher)
Pull request description:
This is joint work with @ mzumsande.
note: this requires a pruned node with deep reorgs to trigger. still it breaks assumptions in the codebase and is good to fix. A similar UB was fixed in https://github.com/bitcoin/bitcoin/pull/34521.
This PR prevents duplicate insertions into `m_blocks_unlinked` in `FindMostWorkChain`. There are 3 ways to insert into `m_blocks_unlinked`:
1. `LoadBlockIndex` - not problematic, as each block index is processed only once.
2. `ReceivedBlockTransactions` - not problematic, as this is usually only called once per block when it is first accepted in `AcceptBlock`. in the rare case it’s triggered again after pruning, the block would have been removed from `m_blocks_unlinked` when it was initially pruned, so duplicates still can’t arise.
3. `FindMostWorkChain` - problematic when multiple candidate tips share common chains of ancestors, traversals from each tip to the fork point may insert duplicate (`pprev`, `pindex`) entries for blocks whose parents have been pruned.
When the missing parent is later received and `ReceivedBlockTransactions` processes `m_blocks_unlinked`, the same entry may be processed multiple times. This can result in the block being re-added to `setBlockIndexCandidates` with a modified `nSequenceId`, violating its ordering invariants and leading to undefined behavior. So avoid duplicate insertions into `m_blocks_unlinked` in `FindMostWorkChain`.
### how to test:
use the updated `feature_pruning.py` which adds coverage for this scenario.
- on master: the test (with the below diff) fails since `nSequenceId` is being modified for an entry in `setBlockIndexCandidates`
- on this branch: the test (with the below diff) passes
```diff --git a/src/validation.cpp b/src/validation.cpp
--- a/src/validation.cpp
+++ b/src/validation.cpp
@@ -3814,6 +3814,12 @@ void ChainstateManager::ReceivedBlockTransactions(const CBlock& block, CBlockInd
pindex->nHeight, pindex->m_chain_tx_count, prev_tx_sum(*pindex), CLIENT_NAME, FormatFullVersion(), CLIENT_BUGREPORT);
}
pindex->m_chain_tx_count = prev_tx_sum(*pindex);
+ for (const auto& c : m_chainstates) {
+ if (c->setBlockIndexCandidates.contains(pindex)) {
+ LogInfo("### pindex UB = %s", pindex);
+ assert(false);
+ }
+ }
pindex->nSequenceId = nBlockSequenceId++;
for (const auto& c : m_chainstates) {
c->TryAddBlockIndexCandidate(pindex);
```
ACKs for top commit:
sedited:
Re-ACK d6359937bf
marcofleon:
crACK d6359937bf
stringintech:
ACK d6359937bf
mzumsande:
Sure - Code Review ACK [d635993](d6359937bf)
Tree-SHA512: bb21adc2d92fe1865bbbcebf775a850ca3eccac6fe83d7bca10b78eee4c0abf782e44fa0ddfec9d9a70f42fa40bdc49b68a1d1b4905cdc371ba29117d3120619
Creates logic to deprecate logging categories but still
"support" them so the software doesn't quit with unknown category
on startup. Deprecated categories are always false and attempts
to switch them are logged as warnings.
`feature_reindex.py` mines a long chain before restarting during reindex.
Bump its local RPC timeout to avoid client timeouts in parallel debug-mode runs.
fab2874269 lint: Require scripted-diff script to succeed (MarcoFalke)
Pull request description:
Currently, scripted diffs may silently pass with errors.
Fix this issue by calling the script from a Bash instance with error checking enabled: `bash -o errexit -o nounset -o pipefail -c "$SCRIPT"`.
Also, use Bash (not sh) when launching the script itself, because Bash is required anyway.
Can be tested by running something like this and observing the behavior before and after:
```
git commit --allow-empty -m $'scripted-diff: foo\n\n-BEGIN VERIFY SCRIPT-\n false;falseasfsafsaf;true;false|cat; echo "${NO_UN_SET}"|cat \n-END VERIFY SCRIPT-\n' && ./test/lint/commit-script-check.sh HEAD~..HEAD ; echo $?
```
Alternatively, an ancient brittle script can be tested:
```
./test/lint/commit-script-check.sh fb65dde147f63422c4148b089c2f5be0bf5ba80f~..fb65dde147f63422c4148b089c2f5be0bf5ba80f
ACKs for top commit:
hodlinator:
ACK fab2874269
sedited:
ACK fab2874269
Tree-SHA512: e3e8167e150be45a096d4883057640eb5624456f21b134cfa901fe490e5afb192855e55a752cb6121399314f65db155544a60645a101c39a084a62de4af23298
735b25519a support: clamp RLIMIT_MEMLOCK to size_t (Sjors Provoost)
8ab4b9fc85 init: clamp fd limits to int (Sjors Provoost)
4afbabdcef Fix startup failure with RLIM_INFINITY fd limits (Sjors Provoost)
Pull request description:
When setting the fd limit to unlimited, the node fails to start:
```sh
ulimit -n unlimited
build/bin/bitcoind
Error: Not enough file descriptors available. -1 available, 160 required.
```
This was caused by `RaiseFileDescriptorLimit()` (introduced in #2568) casting `limitFD.rlim_cur` to `int`, which for `RLIM_INFINITY` overflows to `-1`. Fix it by returning `std::numeric_limits<int>::max()` instead.
Some platforms implement `RLIM_INFINITY` as the maximum uint64, others as int64 (-1). So simply changing the return type to `uint64_t` wouldn't work.
Similarly, though unlikely to actually happen:
```sh
ulimit -n 214748364
build/bin/bitcoind
Error: Not enough file descriptors available. -2147483648 available, 160 required.
```
The second commit expands the fix by clamping all values above `std::numeric_limits<int>::max()` instead of letting them overflow.
This PR also expands `test/functional/feature_init.py` to cover these, using `resource.setrlimit`. The check is skipped on environments with a hard limit below infinity (or that don't have the Python [Resource module](https://docs.python.org/3/library/resource.html)).
macOS by default has a hard limit of infinity, but on e.g. Ubuntu the default hard limit is 524288.
The third commit applies a similar fix to `PosixLockedPageAllocator::GetLimit()` for 32-bit systems, but without a test.
ACKs for top commit:
winterrdog:
Re-ACK 735b25519a
achow101:
ACK 735b25519a
sedited:
Re-ACK 735b25519a
pinheadmz:
ACK 735b25519a
Tree-SHA512: 0ce0292ecd61456bdec6943b06cbb9ecfc5180ee6dce850f8496ef54af22c1fae6ea473085202f5ba6f72e4dc51a29247620c9a0eae31e96658adc77b293129f
0654511e1b util: Check write failures before renaming settings.json (Shrey)
Pull request description:
This PR Fixes#35373.
The error message was updated from: "This is probably caused by disk corruption or a crash" to: "This is probably caused by a full disk, disk corruption or a crash"
The following files were updated:
1. src/common/settings.cpp -- Updated the main string formatting for the parse failure.
2. src/test/settings_tests.cpp -- Updated the exact string expectation in the C++ unit test.
3. test/functional/feature_settings.py -- Updated the expected error message in the Python test suite.
Configured the Bitcoin Core project natively on Windows using Visual Studio (cmake --preset vs2026-static -DBUILD_GUI=OFF) and it compiled with 100% of the C++ tests passed successfully (347 out of 347).
ACKs for top commit:
maflcko:
review ACK 0654511e1b💧
winterrdog:
ACK 0654511e1b
sedited:
ACK 0654511e1b
ryanofsky:
Code review ACK 0654511e1b with commit message and error message improved since last review
Tree-SHA512: d7b49860f2081a5a9d1d44917b0cf372a77f10cf21773c9b3291871c788a122ec5dde063fdf9e1052cf45b8d2667e15585b952e9f037e84f98fb41546bdd1fa9
Using these classes allows the developers to dynamically create
xprvs and xpubs so that they don't need to hardcode such long keys
in the tests that most of the times clutter the tests and make them
difficult to update.
Co-authored-by: w0xlt <94266259+w0xlt@users.noreply.github.com>
This commit also changes the default sendcmpct version in the functional
test to `2`, since this is the version that nodes expect, prior to this
commit, nowhere in the functional test framework was the default version
value used:
git grep -P 'msg_sendcmpct\((?![^)]*version\s*=)' HEAD^
`version=2` is a sensible default, since this is the version nodes
currently expect in the SENDCMPCT handshake.
blocksonly nodes don't benefit from compact blocks, since they don't
have a mempool to aid in reconstruction, so they should not process
CMPCTBLOCK messages.
This is not just belt-and-suspenders, as a blocksonly node will
trivially reveal exactly which transactions in a block are its own in
the GETBLOCKTXN response to a CMPCTBLOCK. Since it will be missing every
transaction in the block, except for its own.
See discussion: https://github.com/bitcoin/bitcoin/issues/28272
Modifies the invalid_cmpctblock_message test to check that both HB peers
sending unsolicited and non-HB peers sending solicited invalid
cmpctblock's get the boot from us. Also refactors the test to make it
less stateful.
Processing unsolicited CMPCTBLOCK's from a peer that has not been marked
high bandwidth is not well-specified behavior in BIP-0152, in fact the
BIP seems to imply that it is not permitted:
"[...] method is not useful for compact blocks because `cmpctblock`
blocks can be sent unsolicitedly in high-bandwidth mode"
See https://github.com/bitcoin/bips/blob/master/bip-0152.mediawiki#separate-version-for-segregated-witness
This partially mitigates a mempool leak described in
[#28272](https://github.com/bitcoin/bitcoin/issues/28272), but that
particular issue will persist for peers that have been selected as high
bandwidth.
This also mitigates potential DoS / bandwidth-wasting / abusive
behavior that is discussed in the comments of #28272.
Clear the getblocktxn message so we're not checking existing messages,
and check that the hash in the getblocktxn match the cmpctblock being
announced.
Without this tightening of checks, a later commit that ignores
CMPCTBLOCK messages that are unsolicited will succeed these tests while
silently failing in reality.
Co-authored-by: Hodlinator <172445034+hodlinator@users.noreply.github.com>
