The previous diff touched most files in ./test/, so bump the headers to
avoid having to touch them again for a bump later.
-BEGIN VERIFY SCRIPT-
./contrib/devtools/copyright_header.py update ./test/
-END VERIFY SCRIPT-
and update the function name to CheckAddrman (drop "Force") for
nicer log output as it is prefixed to each of these log messages:
2021-09-21T18:42:50Z [opencon] CheckAddrman: new 64864, tried 1690, total 66554 started
2021-09-21T18:42:50Z [opencon] CheckAddrman: completed (76.21ms)
The existing Doxygen documentation on the function already makes
clear that it is unaffected by m_consistency_check_ratio.
This is the same approach as for the addpeeraddress test in
`test/functional/rpc_net.py` in commit 869f1368.
The probability of collision when adding an addrman entry is
expected to be 1/2^16 = 1/65536 for an address from a different /16.
This change hopes to avoid these collisions by adding 1 tried entry
before adding 1 new table one, instead of 2 tried entries followed
by 2 new entries, which appears to have caused a collision in the CI.
To verify the regression test stills fails when expected:
- git checkout 181a120 && git cherry-pick ef242f5
- recompile bitcoind
- git checkout this branch and run test/functional/feature_asmap.py. Expected output:
```
AssertionError: Unexpected stderr bitcoind: ./addrman.h:739: void CAddrMan::Check() const: Assertion `false' failed. !=
```
Co-authored-by: Martin Zumsande <mzumsande@gmail.com>
PR #22697 introduced a reproducible issue in commit 181a1207 that causes the
addrman tried table to fail consistency checks and significantly lose peer
entries when the `-asmap` configuration option is used.
The issue occurs on bitcoind restart due to an initialization order change
in `src/init.cpp` in that commit whereby CAddrman asmap is set after
deserializing `peers.dat`, rather than before.
Issue reported on the `#bitcoin-core-dev` IRC channel starting at
https://www.erisian.com.au/bitcoin-core-dev/log-2021-08-23.html#l-263.
```
addrman lost 22813 new and 2 tried addresses due to collisions or invalid addresses
ADDRMAN CONSISTENCY CHECK FAILED!!! err=-17
bitcoind: ./addrman.h:707: void CAddrMan::Check() const: Assertion `false' failed. Aborted
```
How to reproduce:
- `git checkout 181a1207` and recompile
- launch bitcoind with `-asmap` and `-checkaddrman=1` config options
- restart bitcoind
- bitcoind aborts on second call to `CAddrMan::Check()`
This commit adds a regression test to reproduce the case; it passes or fails
with the same error.
Co-authored-by: John Newbery <john@johnnewbery.com>
Co-authored-by: Martin Zumsande <mzumsande@gmail.com>
and update feature_asmap.py and test_runner.py
This commit moves the asmap init.cpp code from the end of "Step 12: start node"
to "Step 6: network initialization" to provide feedback on passing an -asmap
config arg much more quickly. This change speeds up the feature_asmap.py
functional test file from 60 to 5 seconds by accelerating the 2 tests that use
`assert_start_raises_init_error`.
Credit to Wladimir J. van der Laan for the suggestion.
- move asmap #includes to sorted positions in addrman and init (move-only)
- remove redundant quotes in asmap InitError, update test
- remove full stops from asmap logging to be consistent with debug logging,
update tests
to verify node behaviour and debug log when launching bitcoind in these cases:
1. `bitcoind` with no -asmap arg, using /16 prefix for IP bucketing
2. `bitcoind -asmap=<relative path>`, using the unit test skeleton asmap
3. `bitcoind -asmap/-asmap=` with no file specified, using the default asmap
4. `bitcoind -asmap` with no file specified, and a missing default asmap file
The tests are order-independent. The slowest test (missing default asmap file)
is placed last.
Fabian Jahr