223a4aa [build] Don't fail when passed --disable-lcov and lcov isn't available (fanquake)
Pull request description:
Fixes#10828
As pointed out in #10828, failing with "lcov not found" when we've been passed --disable-lcov doesn't make sense. Master currently behaves like this (where lcov isn't available):
```
./configure --disable-lcov
checking for pkg-config... /usr/local/bin/pkg-config
checking pkg-config is at least version 0.9.0... yes
configure: error: "lcov testing requested but lcov not found"
```
cc @janstary
Tree-SHA512: 606fdbddae67e72fff175f2f34e2c9af4e6972d40d5e1ec5c5d8be5051a728e5b16c35cfd856da0c0ce81dcab9db154a4937b1a6ca1e0233b6e160f2f4362002
Move to AppInitServers. This doesn't have any effects on bitcoin behavior. It
was just strange to have this unrelated code in the middle or parameter
interaction.
6262915 Add unit test for stale tip checking (Suhas Daftuar)
83df257 Add CConnmanTest to mutate g_connman in tests (João Barbosa)
ac7b37c Connect to an extra outbound peer if our tip is stale (Suhas Daftuar)
db32a65 Track tip update time and last new block announcement from each peer (Suhas Daftuar)
2d4327d net: Allow connecting to extra outbound peers (Suhas Daftuar)
Pull request description:
This is an alternative approach to #11534. Rather than disconnect an outbound peer when our tip looks stale, instead try to connect to an additional outbound peer.
Periodically, check to see if we have more outbound peers than we target (ie if any extra peers are in use), and if so, disconnect the one that least recently announced a new block (breaking ties by choosing the newest peer that we connected to).
Tree-SHA512: 8f19e910e0bb36867f81783e020af225f356451899adfc7ade1895d6d3bd5afe51c83759610dfd10c62090c4fe404efa0283b2f63fde0bd7da898a1aaa7fb281
97932cd rpc: further constrain the libevent workaround (Cory Fields)
6b58360 rpc: work-around an upstream libevent bug (Cory Fields)
Pull request description:
A rare race condition may trigger while awaiting the body of a message.
This may fix some reported rpc hangs/crashes.
This work-around mimics what libevent does internally once a write has started, which is what usually happens, but not always due to the processing happening on a different thread: e7ff4ef2b4/http.c (L373)
Fixed upstream at: 5ff8eb2637
Tree-SHA512: b9fa97cae9da2a44101c5faf1e3be0b9cbdf722982d35541cf224be31430779c75e519c8ed18d06ab7487bfb1211069b28f22739f126d6c28ca62d3f73b79a52
720d9e8fa [Wallet] always show help-line of wallet encryption calls (Jonas Schnelli)
Pull request description:
We do currently show/hide the wallet encryption RPC calls from the help if the current wallet.
In case of an encrypted wallet, `encryptwallet` is hidden and `walletpassphrasechange`, `walletpassphrasechange` and `walletlock` do appear in the help.
This is no longer ideal in case of multiwallet due to the fact that one may want help infos in order to target a specific wallet.
IMO its preferable to have a static help screen (show everything always). The currently show/hidden calls do handle the possible invalid encryption-state fine.
Fixes#11588
Tree-SHA512: 513fecd15248a31361f5143685e8cdeb63dfd3fa7120828917e1db54d936dc3db60d48ce46efa5c3a563a48157fe962689879856eeeed53f904686b12aec204e
If our tip hasn't updated in a while, that may be because our peers are
not relaying blocks to us that we would consider valid. Allow connection
to an additional outbound peer in that circumstance.
Also, periodically check to see if we are exceeding our target number of
outbound peers, and disconnect the one which has least recently
announced a new block to us (choosing the newest such peer in the case
of tie).
A rare race condition may trigger while awaiting the body of a message, see
upsteam commit 5ff8eb26371c4dc56f384b2de35bea2d87814779 for details.
This may fix some reported rpc hangs/crashes.
5d465e396 Ensure backupwallet fails when attempting to backup to source file (Tomas van der Wansem)
Pull request description:
Previous behaviour was to destroy the wallet (to zero-length)
This fixes#11375
Tree-SHA512: bfd1738659b15e3f23b6bbdf55ec12269c62c820bf701daec19500b52bd5845bb5516733c6f76f36197eb155182a8a35dc239ad4de2ef1e59bbb0f124a455759
f3d4adf Make p2p-acceptablock not an extended test (Matt Corallo)
00dcda6 [qa] test that invalid blocks on an invalid chain get a disconnect (Matt Corallo)
015a525 Reject headers building on invalid chains by tracking invalidity (Matt Corallo)
932f118 Accept unrequested blocks with work equal to our tip (Matt Corallo)
3d9c70c Stop always storing blocks from whitelisted peers (Matt Corallo)
3b4ac43 Rewrite p2p-acceptblock in preparation for slight behavior changes (Matt Corallo)
Pull request description:
@sdaftuar pointed out that the version in #11487 was somewhat DoS-able as someone could feed you a valid chain that forked off the the last checkpoint block and force you to do lots of work just walking backwards across blocks for each new block they gave you. We came up with a few proposals but settled on the one implemented here as likely the simplest without obvious DoS issues. It uses our existing on-load mapBlockIndex walk to make sure everything that descends from an invalid block is marked as such, and then simply caches blocks which we attempted to connect but which were found to be invalid. To avoid DoS issues during IBD, this will need to depend on #11458.
Includes tests from #11487.
Tree-SHA512: 46aff8332908e122dae72ceb5fe8cd241902c2281a87f58a5fb486bf69d46458d84a096fdcb5f3e8e07fbcf7466232b10c429f4d67855425f11b38ac0bf612e1
b296bf1 Init: Remove redundant exit(EXIT_FAILURE) instances and replace with return false (donaloconnor)
Pull request description:
While reviewing the bitcoin code I noticed that there are a few exit(EXIT_FAILURE) at various places in the AppInit function.
This function returns to main() which will return/exit with EXIT_FAILURE so returning false instead of an explicit exit(EXIT_FAILURE) seems to be cleaner.
This PR attempts to make things a bit more consistent.
There is a subtle difference between exit() and return from main in that the exit() will not clean up any local vars but I don't think this makes a difference in this case. Using exit() might even lead to bugs in the future where the dtor of local objects are expected to be called.
Tree-SHA512: 7d104c3a752b4e7d7bc2382ef7e62543462988f1bbf13dd4077fbeff5399729b76c71a4352556f188b8d306604232477466f5bb827b58a6f3f6273f2370e1faa
f927ee1 Fixed a couple small grammatical errors. (Christian Gentry)
Pull request description:
1. "If a pull request is not to be considered for merging (yet), please
prefix the ..."
2. If a particular commit references another issue, please add the reference. For
example: `refs #1234` or `fixes #4321`.
Tree-SHA512: b2ed11a235800a6b8e9450937352954a2222eb6f08f9556c8f298fd3d64d18e731397b46f3141eab01e0196f53fa3a9d84fb707a1e7691a63dd146b3c5298fe5
659b206 Make listsinceblock refuse unknown block hash (Russell Yanofsky)
Pull request description:
Change suggested by @theuni who noticed listsinceblock would ignore invalid block hashes causing it to return a completely unfiltered list of transactions.
Tree-SHA512: 3c8fb160265780d1334e856e853ab48e2e18372b8f1fc71ae480c3f45317048cc1fee0055d5c58031981a91b9c2bdbeb8e49a889d04ecba61729ce8109f2ce3f
60b98f8 [Util] Update tinyformat.h (fanquake)
Pull request description:
Updates `tinyformat.h` to commit c42f/tinyformat@689695c upstream. Including:
8a2812d8485d9e05a34748e2e48789@achow101 mentioned that since upgrading to Ubuntu 17.10 (GCC 7), tinyformat had been throwing lots of -Wimplicit-fallthrough warnings. However fallthrough warnings should have been silenced by #10489. cc @theuni.
The upstream commit to fix fallthrough warnings is in this PR https://github.com/c42f/tinyformat/pull/39.
The last time tinyformat.h was updated in this repo was in #8274.
Tree-SHA512: a51bd30544693550e08148daf5d244e3a3a410caff7897351eb9cd28f661dc85e193e045bb86068ee4006b2f89a7233b7573b8c50d93d2a9a15a11386fdcc605
9d30f54 [Docs] Update OpenBSD Build Instructions for OpenBSD 6.2 (fanquake)
Pull request description:
This updates the OpenBSD build docs to reflect building [master](8ddf60db7a) on a OpenBSD 6.2 VM (using VirtualBox 5.1.28 r117968 on macOS 10.12.6).
Versions of installed packages were:
```
gmake 4.2.1
g++ 4.9.4
git 2.12.2
libevent 2.0.22
libtool 2.4.2
autoconf 2.69p2
automake 1.15p0
python 3.6.0
boost 1.58.0p3
llvm 4.0.0p2
```
The boost package installed via pkg_add now seems to work correctly. So we shouldn't require manual building + patching.
I also wasn't required to make adjustments to any resource limits.
Building with g++ and Clang was successful, using:
```
./configure --disable-wallet --with-gui=no CC=egcc CXX=eg++ CPP=ecpp
```
and
```
./configure --disable-wallet --with-gui=no CC=clang CXX=clang++
```
Running ``` make check ``` worked for ```test/test_bitcoin``` but ```test/util/bitcoin-util-test.py``` failed with:
```
Running test/util/bitcoin-util-test.py...
../test/util/bitcoin-util-test.py
env: python3: No such file or directory
```
So that seems like a configuration issue, Python 3.6 is installed.
Still todo:
- [ ] Check if a manual installation of Berkeley DB is required
- [x] Fix running ```test/util/bitcoin-util-test.py```
- [x] Have someone else verify building
cc @laanwj
Tree-SHA512: 34b176de4865b36dab9d66e74a15c37152e4b6c9784152c30dabbb515d6d9ae9cdbdc7a7b4d777876f91269a6a78cc277ec87775fc6c17dd509f7cf46e89a2b3
This tracks the set of all known invalid-themselves blocks (ie
blocks which we attempted to connect but which were found to be
invalid). This is used to cheaply check if new headers build on an
invalid chain.
While we're at it we also resolve an edge-case in invalidateblock
on pruned nodes which results in them needing a reindex if they
fail to reorg.
This is a simple change that makes our accept requirements the
same as our request requirements, (ever so slightly) further
decoupling our consensus logic from our FindNextBlocksToDownload
logic in net_processing.
There is no reason to wish to store blocks on disk always just
because a peer is whitelisted. This appears to be a historical
quirk to avoid breaking things when the accept limits were added.
Removes checking whitelisted behavior (which will be removed, the
difference in behavior here makes little sense) and no longer
requires that blocks at the same work as our tip be dropped if not
requested (in part because we *do* request those blocks).
2530bf2 net: Add missing lock in ProcessHeadersMessage(...) (practicalswift)
Pull request description:
Add missing lock in `ProcessHeadersMessage(...)`.
Reading the variable `mapBlockIndex` requires holding the mutex `cs_main`.
The new "Disconnect outbound peers relaying invalid headers" code added in commit 37886d5e2f and merged as part of #11568 two days ago did not lock `cs_main` prior to accessing `mapBlockIndex`.
Tree-SHA512: b799c234be8043d036183a00bc7867bbf3bd7ffe3baa94c88529da3b3cd0571c31ed11dadfaf29c5b8498341d6d0a3c928029a43b69f3267ef263682c91563a3
Nowhere else in the protocol do we send headers which are for
blocks we have not fully validated except in response to getheaders
messages with a null locator. On my public node I have not seen any
such request (whether for an invalid block or not) in at least two
years of debug.log output, indicating that this should have minimal
impact.
Reading the variable mapBlockIndex requires holding the mutex cs_main.
The new "Disconnect outbound peers relaying invalid headers" code
added in commit 37886d5e2f and merged
as part of #11568 two days ago did not lock cs_main prior to accessing
mapBlockIndex.
Warnings prior to this commit:
```
addrman.cpp:390:24: warning: comparison of integers of different signs: 'size_type' (aka 'unsigned long') and 'int' [-Wsign-compare]
if (vRandom.size() != nTried + nNew)
~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~
addrman.cpp:411:52: warning: comparison of integers of different signs: 'int' and 'size_type' (aka 'unsigned long') [-Wsign-compare]
if (info.nRandomPos < 0 || info.nRandomPos >= vRandom.size() || vRandom[info.nRandomPos] != n)
~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~
addrman.cpp:419:25: warning: comparison of integers of different signs: 'size_type' (aka 'unsigned long') and 'int' [-Wsign-compare]
if (setTried.size() != nTried)
~~~~~~~~~~~~~~~ ^ ~~~~~~
addrman.cpp:421:23: warning: comparison of integers of different signs: 'size_type' (aka 'unsigned long') and 'int' [-Wsign-compare]
if (mapNew.size() != nNew)
~~~~~~~~~~~~~ ^ ~~~~
4 warnings generated.
```
cc5c39d [Build] Add AM_OBJCXXFLAGS and QT_PIE_FLAGS to OBJCXXFLAGS to future-proof darwin targets (fanquake)
f8c6697 Fix automake warnings when running autogen.sh (Evan Klitzke)
Pull request description:
Adjusted @eklitzke's commit to completely remove GZIP_ENV.
Added a commit to address OBJCXXFLAGS.
Rebased on master.
Relevant info from @theuni & #11013 below.
--------
GZIP_ENV was indeed added for determinism, but gitian exports this as needed, so it's not really necessary. I'd rather just remove it.
The mm.o rule was added to support XCode 4.2's ancient version of automake. That's irrelevant now, so it makes sense to remove that too.
All darwin targets are PIE by default, so we don't technically need the flags, but I'd be more comfortable if we hooked up the OBJCXXFLAGS in case future ones are added.
--------
The second commit addresses the last point, but could probably use a better commit message.
These warnings are removed from autogen output:
```
Makefile.am:12: warning: user variable 'GZIP_ENV' defined here ...
/usr/local/Cellar/automake/1.15.1/share/automake-1.15/am/distdir.am: ... overrides Automake variable 'GZIP_ENV' defined here
src/Makefile.am: installing 'build-aux/depcomp'
src/Makefile.am:503: warning: user target '.mm.o' defined here ...
/usr/local/Cellar/automake/1.15.1/share/automake-1.15/am/depend2.am: ... overrides Automake target '.mm.o' defined here
```
Tree-SHA512: bd59df5f6d3aafe35d5e36925bfe61cc71e774583a0438d7dd946c9e7ecf6e59d42f90a58b8cfef0faa404c81050338ad4cefe721b4a949af881e73b6ab254d4
37886d5e2 Disconnect outbound peers relaying invalid headers (Suhas Daftuar)
4637f1852 moveonly: factor out headers processing into separate function (Suhas Daftuar)
Pull request description:
Alternate to #11446.
Disconnect outbound (non-manual) peers that serve us block headers that are already known to be invalid, but exempt compact block announcements from such disconnects.
We restrict disconnection to outbound peers that are using up an outbound connection slot, because we rely on those peers to give us connectivity to the honest network (our inbound peers are not chosen by us and hence could all be from an attacker/sybil). Maintaining connectivity to peers that serve us invalid headers is sometimes desirable, eg after a soft-fork, to protect unupgraded software from being partitioned off the honest network, so we prefer to only disconnect when necessary.
Compact block announcements are exempted from this logic to comply with BIP 152, which explicitly permits nodes to relay compact blocks before fully validating them.
Tree-SHA512: 3ea88e4ccc1184f292a85b17f800d401d2c3806fefc7ad5429d05d6872c53acfa5751e3df83ce6b9c0060ab289511ed70ae1323d140ccc5b12e3c8da6de49936
1. "If a pull request is not to be considered for merging (yet), please
prefix the ..."
2. If a particular commit references another issue, please add the reference. For
example: `refs #1234` or `fixes #4321`.
fd3a2f3 [tests] Add fuzz testing for BlockTransactions and BlockTransactionsRequest (practicalswift)
Pull request description:
The `BlockTransactions` deserialization code is reachable with tainted data via `ProcessMessage(…, "BLOCKTXN", vRecv [tainted], …)`.
The same thing applies to `BlockTransactionsRequest` which is reachable via `"GETBLOCKTXN"`.
Tree-SHA512: 64560ea344bc6145b940472f99866b808725745b060dedfb315be400bd94e55399f50b982149645bd7af7ed9935fd28751d7daf0d3f94a8e2ed3bc52e3325ffb
e065249 Add unit test for outbound peer eviction (Suhas Daftuar)
5a6d00c Permit disconnection of outbound peers on bad/slow chains (Suhas Daftuar)
c60fd71 Disconnecting from bad outbound peers in IBD (Suhas Daftuar)
Pull request description:
The first commit will disconnect an outbound peer that serves us a headers chain with insufficient work while we're in IBD.
The second commit introduces a way to disconnect outbound peers whose chains fall out of sync with ours:
For a given outbound peer, we check whether their best known block (which is known from the blocks they announce to us) has at least as much work as our tip. If it doesn't, we set a 20 minute timeout, and if we still haven't heard about a block with as much work as our tip had when we set the timeout, then we send a single getheaders message, and wait 2 more minutes. If after two minutes their best known block has insufficient work, we disconnect that peer.
We protect 4 of our outbound peers (who provide some "good" headers chains, ie a chain with at least as much work as our tip at some point) from being subject to this logic, to prevent excessive network topology changes as a result of this algorithm, while still ensuring that we have a reasonable number of nodes not known to be on bogus chains.
We also don't require our peers to be on the same chain as us, to prevent accidental partitioning of the network in the event of a chain split. Note that if our peers are ever on a more work chain than our tip, then we will download and validate it, and then either reorg to it, or learn of a consensus incompatibility with that peer and disconnect. This PR is designed to protect against peers that are on a less work chain which we may never try to download and validate.
Tree-SHA512: 2e0169a1dd8a7fb95980573ac4a201924bffdd724c19afcab5efcef076fdbe1f2cec7dc5f5d7e0a6327216f56d3828884f73642e00c8534b56ec2bb4c854a656
Currently we have no rotation of outbound peers. If an outbound peer
stops serving us blocks, or is on a consensus-incompatible chain with
less work than our tip (but otherwise valid headers), then we will never
disconnect that peer, even though that peer is using one of our 8
outbound connection slots. Because we rely on our outbound peers to
find an honest node in order to reach consensus, allowing an
incompatible peer to occupy one of those slots is undesirable,
particularly if it is possible for all such slots to be occupied by such
peers.
Protect against this by always checking to see if a peer's best known
block has less work than our tip, and if so, set a 20 minute timeout --
if the peer is still not known to have caught up to a chain with as much
work as ours after 20 minutes, then send a single getheaders message,
wait 2 more minutes, and if a better header hasn't been received by then,
disconnect that peer.
Note:
- we do not require that our peer sync to the same tip as ours, just an
equal or greater work tip. (Doing otherwise would risk partitioning the
network in the event of a chain split, and is also unnecessary.)
- we pick 4 of our outbound peers and do not subject them to this logic,
to be more conservative. We don't wish to permit temporary network
issues (or an attacker) to excessively disrupt network topology.
practicalswift