Several tests call disconnect_nodes() on each node-pair in rapid
succession, resulting in a race condition if a node disconnects a peer
in-between the calculation of the nodeid's to disconnect and the
invocation of the disconnectnode rpc call. Handle this.
GitHub-Pull: #13201
Rebased-From: 09c6699
If a cookie file exists in a datadir prior to node startup, it must have
been leftover from a prior unclean shutdown. As bitcoind will overwrite
it anyway, delete it before starting up to prevent the test framework
from inadvertently trying to connect using stale credentials.
Github-Pull: #12902
Rebased-From: 75d0e4c5444249f0cf81ed0494e2c2717170b695
The output would produce arbitrary number of decimal points, sometimes resulting in 9 decimals:
AssertionError: Fee of 0.00000415 BTC too low! (Should be 0.000006175 BTC)
The above looks like the expected fee is 6175 sats when in reality it's 618.
Github-Pull: #12486
Rebased-From: 42e1b5d9797b65d3ce13a7cbace15fbedbcd4018
TestNode is a class responsible for all state related to a bitcoind node
under test. It stores local state, is responsible for tracking the
bitcoind process and delegates unrecognised messages to the RPC
connection.
This commit changes start_nodes and stop_nodes to start and stop the
bitcoind nodes in parallel, making test setup and teardown much faster.
This commit moves functions start_node, start_nodes, stop_node and
stop_nodes functions into the BitcoinTestFramework class. It also moves
the bitcoind_processes dict and coverage variables into BitcoinTestFramework.
Adds new functional test, dbcrash.py, which uses -dbcrashratio to exercise the
logic for recovering from a crash during chainstate flush.
dbcrash.py is added to the extended tests, as it may take ~10 minutes to run
Use _Exit() instead of exit() for crash simulation
This eliminates stderr output such as:
terminate called without an active exception
or
Assertion failed: (!pthread_mutex_destroy(&m)), function ~recursive_mutex, file /usr/local/include/boost/thread/pthread/recursive_mutex.hpp, line 104.
Eliminating the stderr output on crash simulation allows testing with
test_runner.py, which reports a test as failed if stderr is produced.
This has no effect on the outcome of test cases, but
prints shorter and less confusing tracebacks on fails.
I.e. does not print an obvious "ConnectionRefusedError"
when shutting down the nodes due to an invalid return code.
Better to check that rpcuser and rpcpassword exist then to check for
the cookie in the test framework.
Name an argument for consistency in p2p-segwit.py
This commit marks the start/stop functions in util.py as private module
functions. A future PR will remove these entirely and move the
functionality directly into the BitcoinTestFramework class, but setting them as
private in this PR will prevent anyone from accidentally calling them
before that future PR is merged.
This commit reduces spammy logging by the test framework. It truncates
logging send/receive message in mininode to 500 characters. mininode
was previously logging the entire message sent received, which can be up
to 1MB for a full block.
This commit fixes the module-level docstrings for the tests and helper
modules in qa. Many of these tests were uncommented previously - this
commit ensures that every test case has at least a minimum level of
commenting.
Make sure that RPC tests are actually checking failures correctly by:
- Catching JSON RPC exceptions and verifying the error codes and messages.
- Failing the test case if the JSON RPC exception isn't raised.
Motivation for this change is mainly to make sync_blocks behavior easier to
understand. Behavior is unchanged in the normal case when there are only 2
nodes in the rpc_connections set. When there are more than 2 nodes, the
previous "timeout -= wait" statement wouldn't take into account time spent
waiting for all nodes and as a result could lead to blocking for longer than
the requested timeout.
Suhas Daftuar