Merge 882375f2eb3942be644591ae45fbe57458ed40ee into 5f4422d68dc3530c353af1f87499de1c864b60ad

3301d2cbe8c3b76c97285d75fa59637cb6952d0b qa: Wait for txindex to avoid race condition (Hodlinator)
9bfb0d75ba10591cc6c9620f9fd1ecc0e55e7a48 qa: Remove unnecessary -txindex args (Hodlinator)
7ac281c19cd3d11f316dbbb3308eabf1ad4f26d6 qa: Add missing coverage of corrupt indexes (Hodlinator)

Pull request description:

  - Add synchronization in 3 places where if the Transaction Index happens to be slow, we get rare test failures when querying it for transactions (one such case experienced on Windows, prompting investigation).
  - Remove unnecessary TxIndex initialization in some tests.
  - Add some test coverage where TxIndex aspect could be tested in feature_init.py.

ACKs for top commit:
  fjahr:
    re-ACK 3301d2cbe8c3b76c97285d75fa59637cb6952d0b
  mzumsande:
    Code Review ACK 3301d2cbe8c3b76c97285d75fa59637cb6952d0b
  furszy:
    Code review ACK 3301d2cbe8c3b76c97285d75fa59637cb6952d0b
  Prabhat1308:
    Concept ACK [`3301d2c`](3301d2cbe8)

Tree-SHA512: 7c2019e38455f344856aaf6b381faafbd88d53dc88d13309deb718c1dcfbee4ccca7c7f1b66917395503a6f94c3b216a007ad432cc8b93d0309db9805f38d602

doc/zmq.md

@@ -92,35 +92,45 @@ corresponds to the notification type. For instance, for the
 notification `-zmqpubhashtx` the topic is `hashtx` (no null
 terminator). These options can also be provided in bitcoin.conf.
 
-The topics are:
+All ZMQ messages are multipart messages that share the same structure with three parts:
 
-`sequence`: the body is structured as the following based on the type of message:
+    | topic | body | <uint32 sequence number in Little Endian> |
 
-    <32-byte hash>C :                 Blockhash connected
-    <32-byte hash>D :                 Blockhash disconnected
-    <32-byte hash>R<8-byte LE uint> : Transactionhash removed from mempool for non-block inclusion reason
-    <32-byte hash>A<8-byte LE uint> : Transactionhash added mempool
+where the last part is a sequence number (representing the message count to detect lost messages), distinct for each topic.
 
-Where the 8-byte uints correspond to the mempool sequence number.
+**_NOTE:_**  All 32-byte hashes are in _reversed byte order_ (i. e. with bytes produced by hashing function reversed), the same format as the RPC interface and block explorers use to display transaction and block hashes.
 
-`rawtx`: Notifies about all transactions, both when they are added to mempool or when a new block arrives. This means a transaction could be published multiple times. First, when it enters the mempool and then again in each block that includes it. The messages are ZMQ multipart messages with three parts. The first part is the topic (`rawtx`), the second part is the serialized transaction, and the last part is a sequence number (representing the message count to detect lost messages).
+The topics and bodies are:
 
-    | rawtx | <serialized transaction> | <uint32 sequence number in Little Endian>
+`sequence`: The message can have one of the following forms:
 
-`hashtx`: Notifies about all transactions, both when they are added to mempool or when a new block arrives. This means a transaction could be published multiple times. First, when it enters the mempool and then again in each block that includes it. The messages are ZMQ multipart messages with three parts. The first part is the topic (`hashtx`), the second part is the 32-byte transaction hash, and the last part is a sequence number (representing the message count to detect lost messages).
+    | sequence | <32-byte block hash, reversed>C                       | <uint32 sequence number in Little Endian> |
+    | sequence | <32-byte block hash, reversed>D                       | <uint32 sequence number in Little Endian> |
+    | sequence | <32-byte transaction hash, reversed>R<8-byte LE uint> | <uint32 sequence number in Little Endian> |
+    | sequence | <32-byte transaction hash, reversed>A<8-byte LE uint> | <uint32 sequence number in Little Endian> |
 
-    | hashtx | <32-byte transaction hash in Little Endian> | <uint32 sequence number in Little Endian>
+Where the 8-byte uints correspond to the mempool sequence number and the types of bodies are:
 
+   - `C` : block with this hash connected
+   - `D` : block with this hash disconnected
+   - `R` : transaction with this hash removed from mempool for non-block inclusion reason
+   - `A` : transaction with this hash added to mempool
 
-`rawblock`: Notifies when the chain tip is updated. When assumeutxo is in use, this notification will not be issued for historical blocks connected to the background validation chainstate. Messages are ZMQ multipart messages with three parts. The first part is the topic (`rawblock`), the second part is the serialized block, and the last part is a sequence number (representing the message count to detect lost messages).
+`rawtx`: Notifies about all transactions, both when they are added to mempool or when a new block arrives. This means a transaction could be published multiple times. First, when it enters the mempool and then again in each block that includes it. The body part of the message is the serialized transaction:
 
-    | rawblock | <serialized block> | <uint32 sequence number in Little Endian>
+    | rawtx | <serialized transaction> | <uint32 sequence number in Little Endian> |
 
-`hashblock`: Notifies when the chain tip is updated. When assumeutxo is in use, this notification will not be issued for historical blocks connected to the background validation chainstate. Messages are ZMQ multipart messages with three parts. The first part is the topic (`hashblock`), the second part is the 32-byte block hash, and the last part is a sequence number (representing the message count to detect lost messages).
+`hashtx`: Notifies about all transactions, both when they are added to mempool or when a new block arrives. This means a transaction could be published multiple times. First, when it enters the mempool and then again in each block that includes it. The body part of the mesage is the 32-byte transaction hash in reversed byte order:
 
-    | hashblock | <32-byte block hash in Little Endian> | <uint32 sequence number in Little Endian>
+    | hashtx | <32-byte transaction hash, reversed> | <uint32 sequence number in Little Endian> |
 
-**_NOTE:_**  Note that the 32-byte hashes are in Little Endian and not in the Big Endian format that the RPC interface and block explorers use to display transaction and block hashes.
+`rawblock`: Notifies when the chain tip is updated. When assumeutxo is in use, this notification will not be issued for historical blocks connected to the background validation chainstate. The body part of the message is the serialized block:
+
+    | rawblock | <serialized block> | <uint32 sequence number in Little Endian> |
+
+`hashblock`: Notifies when the chain tip is updated. When assumeutxo is in use, this notification will not be issued for historical blocks connected to the background validation chainstate. The body part of the message is the 32-byte block hash in reversed byte order:
+
+    | hashblock | <32-byte block hash, reversed> | <uint32 sequence number in Little Endian> |
 
 ZeroMQ endpoint specifiers for TCP (and others) are documented in the
 [ZeroMQ API](http://api.zeromq.org/4-0:_start).

test/functional/feature_init.py

@@ -88,7 +88,7 @@ class InitTest(BitcoinTestFramework):
 
         args = ['-txindex=1', '-blockfilterindex=1', '-coinstatsindex=1']
         for terminate_line in lines_to_terminate_after:
-            self.log.info(f"Starting node and will exit after line {terminate_line}")
+            self.log.info(f"Starting node and will terminate after line {terminate_line}")
             with node.busy_wait_for_debug_log([terminate_line]):
                 if platform.system() == 'Windows':
                     # CREATE_NEW_PROCESS_GROUP is required in order to be able
@@ -108,12 +108,22 @@ class InitTest(BitcoinTestFramework):
             'blocks/index/*.ldb': 'Error opening block database.',
             'chainstate/*.ldb': 'Error opening coins database.',
             'blocks/blk*.dat': 'Error loading block database.',
+            'indexes/txindex/MANIFEST*': 'LevelDB error: Corruption: CURRENT points to a non-existent file',
+            # Removing these files does not result in a startup error:
+            # 'indexes/blockfilter/basic/*.dat', 'indexes/blockfilter/basic/db/*.*', 'indexes/coinstats/db/*.*',
+            # 'indexes/txindex/*.log', 'indexes/txindex/CURRENT', 'indexes/txindex/LOCK'
         }
 
         files_to_perturb = {
             'blocks/index/*.ldb': 'Error loading block database.',
             'chainstate/*.ldb': 'Error opening coins database.',
             'blocks/blk*.dat': 'Corrupted block database detected.',
+            'indexes/blockfilter/basic/db/*.*': 'LevelDB error: Corruption',
+            'indexes/coinstats/db/*.*': 'LevelDB error: Corruption',
+            'indexes/txindex/*.log': 'LevelDB error: Corruption',
+            'indexes/txindex/CURRENT': 'LevelDB error: Corruption',
+            # Perturbing these files does not result in a startup error:
+            # 'indexes/blockfilter/basic/*.dat', 'indexes/txindex/MANIFEST*', 'indexes/txindex/LOCK'
         }
 
         for file_patt, err_fragment in files_to_delete.items():
@@ -135,9 +145,10 @@ class InitTest(BitcoinTestFramework):
             self.stop_node(0)
 
         self.log.info("Test startup errors after perturbing certain essential files")
+        dirs = ["blocks", "chainstate", "indexes"]
         for file_patt, err_fragment in files_to_perturb.items():
-            shutil.copytree(node.chain_path / "blocks", node.chain_path / "blocks_bak")
-            shutil.copytree(node.chain_path / "chainstate", node.chain_path / "chainstate_bak")
+            for dir in dirs:
+                shutil.copytree(node.chain_path / dir, node.chain_path / f"{dir}_bak")
             target_files = list(node.chain_path.glob(file_patt))
 
             for target_file in target_files:
@@ -151,10 +162,9 @@ class InitTest(BitcoinTestFramework):
 
             start_expecting_error(err_fragment)
 
-            shutil.rmtree(node.chain_path / "blocks")
-            shutil.rmtree(node.chain_path / "chainstate")
-            shutil.move(node.chain_path / "blocks_bak", node.chain_path / "blocks")
-            shutil.move(node.chain_path / "chainstate_bak", node.chain_path / "chainstate")
+            for dir in dirs:
+                shutil.rmtree(node.chain_path / dir)
+                shutil.move(node.chain_path / f"{dir}_bak", node.chain_path / dir)
 
     def init_pid_test(self):
         BITCOIN_PID_FILENAME_CUSTOM = "my_fancy_bitcoin_pid_file.foobar"

test/functional/mempool_accept.py

@@ -45,6 +45,7 @@ from test_framework.util import (
     assert_equal,
     assert_greater_than,
     assert_raises_rpc_error,
+    sync_txindex,
 )
 from test_framework.wallet import MiniWallet
 from test_framework.wallet_util import generate_keypair
@@ -270,6 +271,7 @@ class MempoolAcceptanceTest(BitcoinTestFramework):
 
         self.log.info('A coinbase transaction')
         # Pick the input of the first tx we created, so it has to be a coinbase tx
+        sync_txindex(self, node)
         raw_tx_coinbase_spent = node.getrawtransaction(txid=node.decoderawtransaction(hexstring=raw_tx_in_block)['vin'][0]['txid'])
         tx = tx_from_hex(raw_tx_coinbase_spent)
         self.check_mempool_result(

test/functional/rpc_rawtransaction.py

@@ -34,6 +34,7 @@ from test_framework.util import (
     assert_equal,
     assert_greater_than,
     assert_raises_rpc_error,
+    sync_txindex,
 )
 from test_framework.wallet import (
     getnewdestination,
@@ -70,7 +71,7 @@ class RawTransactionsTest(BitcoinTestFramework):
         self.num_nodes = 3
         self.extra_args = [
             ["-txindex"],
-            ["-txindex"],
+            [],
             ["-fastprune", "-prune=1"],
         ]
         # whitelist peers to speed up tx relay / mempool sync
@@ -109,6 +110,7 @@ class RawTransactionsTest(BitcoinTestFramework):
             self.log.info(f"Test getrawtransaction {'with' if n == 0 else 'without'} -txindex")
 
             if n == 0:
+                sync_txindex(self, self.nodes[n])
                 # With -txindex.
                 # 1. valid parameters - only supply txid
                 assert_equal(self.nodes[n].getrawtransaction(txId), tx['hex'])

test/functional/rpc_txoutproof.py

@@ -12,6 +12,7 @@ from test_framework.test_framework import BitcoinTestFramework
 from test_framework.util import (
     assert_equal,
     assert_raises_rpc_error,
+    sync_txindex,
 )
 from test_framework.wallet import MiniWallet
 
@@ -77,6 +78,7 @@ class MerkleBlockTest(BitcoinTestFramework):
         assert_equal(sorted(self.nodes[0].verifytxoutproof(self.nodes[0].gettxoutproof([txid1, txid2]))), sorted(txlist))
         assert_equal(sorted(self.nodes[0].verifytxoutproof(self.nodes[0].gettxoutproof([txid2, txid1]))), sorted(txlist))
         # We can always get a proof if we have a -txindex
+        sync_txindex(self, self.nodes[1])
         assert_equal(self.nodes[0].verifytxoutproof(self.nodes[1].gettxoutproof([txid_spent])), [txid_spent])
         # We can't get a proof if we specify transactions from different blocks
         assert_raises_rpc_error(-5, "Not all transactions found in specified or retrieved block", self.nodes[0].gettxoutproof, [txid1, txid3])

test/functional/test_framework/util.py

@@ -592,3 +592,10 @@ def find_vout_for_address(node, txid, addr):
         if addr == tx["vout"][i]["scriptPubKey"]["address"]:
             return i
     raise RuntimeError("Vout not found for address: txid=%s, addr=%s" % (txid, addr))
+
+
+def sync_txindex(test_framework, node):
+    test_framework.log.debug("Waiting for node txindex to sync")
+    sync_start = int(time.time())
+    test_framework.wait_until(lambda: node.getindexinfo("txindex")["txindex"]["synced"])
+    test_framework.log.debug(f"Synced in {time.time() - sync_start} seconds")

test/functional/wallet_avoid_mixing_output_types.py

@@ -117,7 +117,6 @@ class AddressInputTypeGrouping(BitcoinTestFramework):
         self.extra_args = [
             [
                 "-addresstype=bech32",
-                "-txindex",
             ],
             [
                 "-addresstype=p2sh-segwit",