simplify advance and reorg

This commit is contained in:
Jack Robison 2021-07-14 13:09:57 -04:00 committed by Victor Shyba
parent 81773a6497
commit acfc1f56ee
3 changed files with 164 additions and 590 deletions

View file

@ -1,12 +1,14 @@
import time import time
import asyncio import asyncio
import typing import typing
import struct
from bisect import bisect_right from bisect import bisect_right
from struct import pack, unpack from struct import pack, unpack
from concurrent.futures.thread import ThreadPoolExecutor from concurrent.futures.thread import ThreadPoolExecutor
from typing import Optional, List, Tuple, Set, DefaultDict, Dict from typing import Optional, List, Tuple, Set, DefaultDict, Dict
from prometheus_client import Gauge, Histogram from prometheus_client import Gauge, Histogram
from collections import defaultdict from collections import defaultdict
import array
import lbry import lbry
from lbry.schema.claim import Claim from lbry.schema.claim import Claim
from lbry.schema.mime_types import guess_stream_type from lbry.schema.mime_types import guess_stream_type
@ -195,21 +197,18 @@ class BlockProcessor:
# Meta # Meta
self.next_cache_check = 0 self.next_cache_check = 0
self.touched = set() self.touched = set()
self.reorg_count = 0
# Caches of unflushed items. # Caches of unflushed items.
self.headers = []
self.block_hashes = [] self.block_hashes = []
self.block_txs = [] self.block_txs = []
self.undo_infos = [] self.undo_infos = []
# UTXO cache # UTXO cache
self.utxo_cache = {} self.utxo_cache: Dict[Tuple[bytes, int], bytes] = {}
self.db_deletes = [] self.db_deletes = []
# Claimtrie cache # Claimtrie cache
self.db_op_stack: Optional[RevertableOpStack] = None self.db_op_stack: Optional[RevertableOpStack] = None
self.undo_claims = []
# If the lock is successfully acquired, in-memory chain state # If the lock is successfully acquired, in-memory chain state
# is consistent with self.height # is consistent with self.height
@ -263,6 +262,7 @@ class BlockProcessor:
self.doesnt_have_valid_signature: Set[bytes] = set() self.doesnt_have_valid_signature: Set[bytes] = set()
self.claim_channels: Dict[bytes, bytes] = {} self.claim_channels: Dict[bytes, bytes] = {}
self.hashXs_by_tx: DefaultDict[bytes, List[int]] = defaultdict(list)
def claim_producer(self): def claim_producer(self):
if self.db.db_height <= 1: if self.db.db_height <= 1:
@ -295,6 +295,7 @@ class BlockProcessor:
"""Process the list of raw blocks passed. Detects and handles """Process the list of raw blocks passed. Detects and handles
reorgs. reorgs.
""" """
if not raw_blocks: if not raw_blocks:
return return
first = self.height + 1 first = self.height + 1
@ -305,7 +306,7 @@ class BlockProcessor:
chain = [self.tip] + [self.coin.header_hash(h) for h in headers[:-1]] chain = [self.tip] + [self.coin.header_hash(h) for h in headers[:-1]]
if hprevs == chain: if hprevs == chain:
start = time.perf_counter() total_start = time.perf_counter()
try: try:
for block in blocks: for block in blocks:
start = time.perf_counter() start = time.perf_counter()
@ -323,14 +324,7 @@ class BlockProcessor:
except: except:
self.logger.exception("advance blocks failed") self.logger.exception("advance blocks failed")
raise raise
# if self.sql: processed_time = time.perf_counter() - total_start
# for cache in self.search_cache.values():
# cache.clear()
self.history_cache.clear() # TODO: is this needed?
self.notifications.notified_mempool_txs.clear()
processed_time = time.perf_counter() - start
self.block_count_metric.set(self.height) self.block_count_metric.set(self.height)
self.block_update_time_metric.observe(processed_time) self.block_update_time_metric.observe(processed_time)
self.status_server.set_height(self.db.fs_height, self.db.db_tip) self.status_server.set_height(self.db.fs_height, self.db.db_tip)
@ -338,13 +332,32 @@ class BlockProcessor:
s = '' if len(blocks) == 1 else 's' s = '' if len(blocks) == 1 else 's'
self.logger.info('processed {:,d} block{} in {:.1f}s'.format(len(blocks), s, processed_time)) self.logger.info('processed {:,d} block{} in {:.1f}s'.format(len(blocks), s, processed_time))
if self._caught_up_event.is_set(): if self._caught_up_event.is_set():
# if self.sql:
# await self.db.search_index.apply_filters(self.sql.blocked_streams, self.sql.blocked_channels,
# self.sql.filtered_streams, self.sql.filtered_channels)
await self.notifications.on_block(self.touched, self.height) await self.notifications.on_block(self.touched, self.height)
self.touched = set() self.touched = set()
elif hprevs[0] != chain[0]: elif hprevs[0] != chain[0]:
await self.reorg_chain() min_start_height = max(self.height - self.coin.REORG_LIMIT, 0)
count = 1
block_hashes_from_lbrycrd = await self.daemon.block_hex_hashes(
min_start_height, self.coin.REORG_LIMIT
)
for height, block_hash in zip(
reversed(range(min_start_height, min_start_height + self.coin.REORG_LIMIT)),
reversed(block_hashes_from_lbrycrd)):
if self.block_hashes[height][::-1].hex() == block_hash:
break
count += 1
self.logger.warning(f"blockchain reorg detected at {self.height}, unwinding last {count} blocks")
try:
assert count > 0, count
for _ in range(count):
await self.run_in_thread_with_lock(self.backup_block)
await self.prefetcher.reset_height(self.height)
self.reorg_count_metric.inc()
except:
self.logger.exception("reorg blocks failed")
raise
finally:
self.logger.info("backed up to block %i", self.height)
else: else:
# It is probably possible but extremely rare that what # It is probably possible but extremely rare that what
# bitcoind returns doesn't form a chain because it # bitcoind returns doesn't form a chain because it
@ -355,101 +368,26 @@ class BlockProcessor:
'resetting the prefetcher') 'resetting the prefetcher')
await self.prefetcher.reset_height(self.height) await self.prefetcher.reset_height(self.height)
async def reorg_chain(self, count: Optional[int] = None):
"""Handle a chain reorganisation.
Count is the number of blocks to simulate a reorg, or None for
a real reorg."""
if count is None:
self.logger.info('chain reorg detected')
else:
self.logger.info(f'faking a reorg of {count:,d} blocks')
async def get_raw_blocks(last_height, hex_hashes):
heights = range(last_height, last_height - len(hex_hashes), -1)
try:
blocks = [await self.db.read_raw_block(height) for height in heights]
self.logger.info(f'read {len(blocks)} blocks from disk')
return blocks
except FileNotFoundError:
return await self.daemon.raw_blocks(hex_hashes)
try:
start, last, hashes = await self.reorg_hashes(count)
# Reverse and convert to hex strings.
hashes = [hash_to_hex_str(hash) for hash in reversed(hashes)]
self.logger.info("reorg %i block hashes", len(hashes))
for hex_hashes in chunks(hashes, 50):
raw_blocks = await get_raw_blocks(last, hex_hashes)
self.logger.info("got %i raw blocks", len(raw_blocks))
await self.run_in_thread_with_lock(self.backup_blocks, raw_blocks)
last -= len(raw_blocks)
await self.prefetcher.reset_height(self.height)
self.reorg_count_metric.inc()
except:
self.logger.exception("boom")
raise
finally:
self.logger.info("done with reorg")
async def reorg_hashes(self, count):
"""Return a pair (start, last, hashes) of blocks to back up during a
reorg.
The hashes are returned in order of increasing height. Start
is the height of the first hash, last of the last.
"""
"""Calculate the reorg range"""
def diff_pos(hashes1, hashes2):
"""Returns the index of the first difference in the hash lists.
If both lists match returns their length."""
for n, (hash1, hash2) in enumerate(zip(hashes1, hashes2)):
if hash1 != hash2:
return n
return len(hashes)
if count is None:
# A real reorg
start = self.height - 1
count = 1
while start > 0:
hashes = await self.db.fs_block_hashes(start, count)
hex_hashes = [hash_to_hex_str(hash) for hash in hashes]
d_hex_hashes = await self.daemon.block_hex_hashes(start, count)
n = diff_pos(hex_hashes, d_hex_hashes)
if n > 0:
start += n
break
count = min(count * 2, start)
start -= count
count = (self.height - start) + 1
else:
start = (self.height - count) + 1
last = start + count - 1
s = '' if count == 1 else 's'
self.logger.info(f'chain was reorganised replacing {count:,d} '
f'block{s} at heights {start:,d}-{last:,d}')
return start, last, await self.db.fs_block_hashes(start, count)
# - Flushing # - Flushing
def flush_data(self): def flush_data(self):
"""The data for a flush. The lock must be taken.""" """The data for a flush. The lock must be taken."""
assert self.state_lock.locked() assert self.state_lock.locked()
return FlushData(self.height, self.tx_count, self.headers, self.block_hashes, return FlushData(self.height, self.tx_count, self.block_hashes,
self.block_txs, self.db_op_stack, self.undo_infos, self.utxo_cache, self.block_txs, self.db_op_stack, self.tip)
self.db_deletes, self.tip, self.undo_claims)
async def flush(self): async def flush(self):
def flush(): def flush():
self.db.flush_dbs(self.flush_data()) self.db.flush_dbs(self.flush_data())
await self.run_in_thread_with_lock(flush) await self.run_in_thread_with_lock(flush)
async def write_state(self):
def flush():
with self.db.db.write_batch() as batch:
self.db.write_db_state(batch)
await self.run_in_thread_with_lock(flush)
def _add_claim_or_update(self, height: int, txo: 'Output', tx_hash: bytes, tx_num: int, nout: int, def _add_claim_or_update(self, height: int, txo: 'Output', tx_hash: bytes, tx_num: int, nout: int,
spent_claims: typing.Dict[bytes, typing.Tuple[int, int, str]]): spent_claims: typing.Dict[bytes, typing.Tuple[int, int, str]]):
try: try:
@ -1167,51 +1105,51 @@ class BlockProcessor:
block_hash = self.coin.header_hash(block.header) block_hash = self.coin.header_hash(block.header)
self.block_hashes.append(block_hash) self.block_hashes.append(block_hash)
self.block_txs.append((b''.join(tx_hash for tx, tx_hash in txs), [tx.raw for tx, _ in txs])) self.db_op_stack.append(RevertablePut(*Prefixes.block_hash.pack_item(height, block_hash)))
first_tx_num = self.tx_count
undo_info = []
hashXs_by_tx = []
tx_count = self.tx_count tx_count = self.tx_count
# Use local vars for speed in the loops # Use local vars for speed in the loops
put_utxo = self.utxo_cache.__setitem__
claimtrie_stash_extend = self.db_op_stack.extend
spend_utxo = self.spend_utxo spend_utxo = self.spend_utxo
undo_info_append = undo_info.append add_utxo = self.add_utxo
update_touched = self.touched.update
append_hashX_by_tx = hashXs_by_tx.append spend_claim_or_support_txo = self._spend_claim_or_support_txo
hashX_from_script = self.coin.hashX_from_script add_claim_or_support = self._add_claim_or_support
for tx, tx_hash in txs: for tx, tx_hash in txs:
spent_claims = {} spent_claims = {}
hashXs = [] # hashXs touched by spent inputs/rx outputs
append_hashX = hashXs.append
tx_numb = pack('<I', tx_count)
txos = Transaction(tx.raw).outputs txos = Transaction(tx.raw).outputs
self.db_op_stack.extend([
RevertablePut(*Prefixes.tx.pack_item(tx_hash, tx.raw)),
RevertablePut(*Prefixes.tx_num.pack_item(tx_hash, tx_count)),
RevertablePut(*Prefixes.tx_hash.pack_item(tx_count, tx_hash))
])
# Spend the inputs # Spend the inputs
for txin in tx.inputs: for txin in tx.inputs:
if txin.is_generation(): if txin.is_generation():
continue continue
txin_num = self.db.transaction_num_mapping[txin.prev_hash]
# spend utxo for address histories # spend utxo for address histories
cache_value = spend_utxo(txin.prev_hash, txin.prev_idx) hashX = spend_utxo(txin.prev_hash, txin.prev_idx)
undo_info_append(cache_value) if hashX:
append_hashX(cache_value[:-12]) # self._set_hashX_cache(hashX)
self._spend_claim_or_support_txo(txin, spent_claims) if txin_num not in self.hashXs_by_tx[hashX]:
self.hashXs_by_tx[hashX].append(txin_num)
# spend claim/support txo
spend_claim_or_support_txo(txin, spent_claims)
# Add the new UTXOs # Add the new UTXOs
for nout, txout in enumerate(tx.outputs): for nout, txout in enumerate(tx.outputs):
# Get the hashX. Ignore unspendable outputs # Get the hashX. Ignore unspendable outputs
hashX = hashX_from_script(txout.pk_script) hashX = add_utxo(tx_hash, tx_count, nout, txout)
if hashX: if hashX:
append_hashX(hashX) # self._set_hashX_cache(hashX)
put_utxo(tx_hash + pack('<H', nout), hashX + tx_numb + pack('<Q', txout.value)) if tx_count not in self.hashXs_by_tx[hashX]:
self.hashXs_by_tx[hashX].append(tx_count)
# add claim/support txo # add claim/support txo
self._add_claim_or_support( add_claim_or_support(
height, tx_hash, tx_count, nout, txos[nout], spent_claims height, tx_hash, tx_count, nout, txos[nout], spent_claims
) )
@ -1220,8 +1158,6 @@ class BlockProcessor:
# print(f"\tabandon {abandoned_claim_hash.hex()} {tx_num} {nout}") # print(f"\tabandon {abandoned_claim_hash.hex()} {tx_num} {nout}")
self._abandon_claim(abandoned_claim_hash, tx_num, nout, name) self._abandon_claim(abandoned_claim_hash, tx_num, nout, name)
append_hashX_by_tx(hashXs)
update_touched(hashXs)
self.db.total_transactions.append(tx_hash) self.db.total_transactions.append(tx_hash)
self.db.transaction_num_mapping[tx_hash] = tx_count self.db.transaction_num_mapping[tx_hash] = tx_count
tx_count += 1 tx_count += 1
@ -1232,31 +1168,34 @@ class BlockProcessor:
# activate claims and process takeovers # activate claims and process takeovers
self._get_takeover_ops(height) self._get_takeover_ops(height)
# self.db.add_unflushed(hashXs_by_tx, self.tx_count) self.db_op_stack.append(RevertablePut(*Prefixes.header.pack_item(height, block.header)))
_unflushed = self.db.hist_unflushed self.db_op_stack.append(RevertablePut(*Prefixes.tx_count.pack_item(height, tx_count)))
_count = 0
for _tx_num, _hashXs in enumerate(hashXs_by_tx, start=first_tx_num): for hashX, new_history in self.hashXs_by_tx.items():
for _hashX in set(_hashXs): if not new_history:
_unflushed[_hashX].append(_tx_num) continue
_count += len(_hashXs) self.db_op_stack.append(
self.db.hist_unflushed_count += _count RevertablePut(
*Prefixes.hashX_history.pack_item(
hashX, height, new_history
)
)
)
self.tx_count = tx_count self.tx_count = tx_count
self.db.tx_counts.append(self.tx_count) self.db.tx_counts.append(self.tx_count)
undo_claims = b''.join(op.invert().pack() for op in self.db_op_stack)
# print("%i undo bytes for %i (%i claimtrie stash ops)" % (len(undo_claims), height, len(claimtrie_stash)))
if height >= self.daemon.cached_height() - self.env.reorg_limit: if height >= self.daemon.cached_height() - self.env.reorg_limit:
self.undo_infos.append((undo_info, height)) self.db_op_stack.append(RevertablePut(*Prefixes.undo.pack_item(height, self.db_op_stack.get_undo_ops())))
self.undo_claims.append((undo_claims, height))
self.db.write_raw_block(block.raw, height)
self.height = height self.height = height
self.headers.append(block.header) self.db.headers.append(block.header)
self.tip = self.coin.header_hash(block.header) self.tip = self.coin.header_hash(block.header)
self.db.flush_dbs(self.flush_data()) self.db.flush_dbs(self.flush_data())
self.clear_after_advance_or_reorg()
def clear_after_advance_or_reorg(self):
self.db_op_stack.clear() self.db_op_stack.clear()
self.txo_to_claim.clear() self.txo_to_claim.clear()
self.claim_hash_to_txo.clear() self.claim_hash_to_txo.clear()
@ -1277,186 +1216,83 @@ class BlockProcessor:
self.expired_claim_hashes.clear() self.expired_claim_hashes.clear()
self.doesnt_have_valid_signature.clear() self.doesnt_have_valid_signature.clear()
self.claim_channels.clear() self.claim_channels.clear()
self.utxo_cache.clear()
# for cache in self.search_cache.values(): self.hashXs_by_tx.clear()
# cache.clear()
self.history_cache.clear() self.history_cache.clear()
self.notifications.notified_mempool_txs.clear() self.notifications.notified_mempool_txs.clear()
def backup_blocks(self, raw_blocks): def backup_block(self):
"""Backup the raw blocks and flush.
The blocks should be in order of decreasing height, starting at.
self.height. A flush is performed once the blocks are backed up.
"""
self.db.assert_flushed(self.flush_data()) self.db.assert_flushed(self.flush_data())
assert self.height >= len(raw_blocks)
coin = self.coin
for raw_block in raw_blocks:
self.logger.info("backup block %i", self.height) self.logger.info("backup block %i", self.height)
# Check and update self.tip # Check and update self.tip
block = coin.block(raw_block, self.height) undo_ops = self.db.read_undo_info(self.height)
header_hash = coin.header_hash(block.header) if undo_ops is None:
if header_hash != self.tip: raise ChainError(f'no undo information found for height {self.height:,d}')
raise ChainError('backup block {} not tip {} at height {:,d}' self.db_op_stack.apply_packed_undo_ops(undo_ops)
.format(hash_to_hex_str(header_hash), self.db_op_stack.append(RevertableDelete(Prefixes.undo.pack_key(self.height), undo_ops))
hash_to_hex_str(self.tip), self.db.headers.pop()
self.height)) self.block_hashes.pop()
self.tip = coin.header_prevhash(block.header)
self.backup_txs(block.transactions)
self.height -= 1
self.db.tx_counts.pop() self.db.tx_counts.pop()
self.tip = self.coin.header_hash(self.db.headers[-1])
while len(self.db.total_transactions) > self.db.tx_counts[-1]:
self.db.transaction_num_mapping.pop(self.db.total_transactions.pop())
self.tx_count -= 1
self.height -= 1
# self.touched can include other addresses which is # self.touched can include other addresses which is
# harmless, but remove None. # harmless, but remove None.
self.touched.discard(None) self.touched.discard(None)
self.db.flush_backup(self.flush_data())
self.db.flush_backup(self.flush_data(), self.touched) self.clear_after_advance_or_reorg()
self.logger.info(f'backed up to height {self.height:,d}') self.logger.info(f'backed up to height {self.height:,d}')
def backup_txs(self, txs): def add_utxo(self, tx_hash: bytes, tx_num: int, nout: int, txout: 'TxOutput') -> Optional[bytes]:
# Prevout values, in order down the block (coinbase first if present)
# undo_info is in reverse block order
undo_info, undo_claims = self.db.read_undo_info(self.height)
if undo_info is None:
raise ChainError(f'no undo information found for height {self.height:,d}')
n = len(undo_info)
# Use local vars for speed in the loops
s_pack = pack
undo_entry_len = 12 + HASHX_LEN
for tx, tx_hash in reversed(txs):
for idx, txout in enumerate(tx.outputs):
# Spend the TX outputs. Be careful with unspendable
# outputs - we didn't save those in the first place.
hashX = self.coin.hashX_from_script(txout.pk_script) hashX = self.coin.hashX_from_script(txout.pk_script)
if hashX: if hashX:
cache_value = self.spend_utxo(tx_hash, idx) self.utxo_cache[(tx_hash, nout)] = hashX
self.touched.add(cache_value[:-12]) self.db_op_stack.extend([
RevertablePut(
# Restore the inputs *Prefixes.utxo.pack_item(hashX, tx_num, nout, txout.value)
for txin in reversed(tx.inputs): ),
if txin.is_generation(): RevertablePut(
continue *Prefixes.hashX_utxo.pack_item(tx_hash[:4], tx_num, nout, hashX)
n -= undo_entry_len )
undo_item = undo_info[n:n + undo_entry_len] ])
self.utxo_cache[txin.prev_hash + s_pack('<H', txin.prev_idx)] = undo_item return hashX
self.touched.add(undo_item[:-12])
self.db.transaction_num_mapping.pop(self.db.total_transactions.pop())
assert n == 0
self.tx_count -= len(txs)
self.undo_claims.append((undo_claims, self.height))
"""An in-memory UTXO cache, representing all changes to UTXO state
since the last DB flush.
We want to store millions of these in memory for optimal
performance during initial sync, because then it is possible to
spend UTXOs without ever going to the database (other than as an
entry in the address history, and there is only one such entry per
TX not per UTXO). So store them in a Python dictionary with
binary keys and values.
Key: TX_HASH + TX_IDX (32 + 2 = 34 bytes)
Value: HASHX + TX_NUM + VALUE (11 + 4 + 8 = 23 bytes)
That's 57 bytes of raw data in-memory. Python dictionary overhead
means each entry actually uses about 205 bytes of memory. So
almost 5 million UTXOs can fit in 1GB of RAM. There are
approximately 42 million UTXOs on bitcoin mainnet at height
433,000.
Semantics:
add: Add it to the cache dictionary.
spend: Remove it if in the cache dictionary. Otherwise it's
been flushed to the DB. Each UTXO is responsible for two
entries in the DB. Mark them for deletion in the next
cache flush.
The UTXO database format has to be able to do two things efficiently:
1. Given an address be able to list its UTXOs and their values
so its balance can be efficiently computed.
2. When processing transactions, for each prevout spent - a (tx_hash,
idx) pair - we have to be able to remove it from the DB. To send
notifications to clients we also need to know any address it paid
to.
To this end we maintain two "tables", one for each point above:
1. Key: b'u' + address_hashX + tx_idx + tx_num
Value: the UTXO value as a 64-bit unsigned integer
2. Key: b'h' + compressed_tx_hash + tx_idx + tx_num
Value: hashX
The compressed tx hash is just the first few bytes of the hash of
the tx in which the UTXO was created. As this is not unique there
will be potential collisions so tx_num is also in the key. When
looking up a UTXO the prefix space of the compressed hash needs to
be searched and resolved if necessary with the tx_num. The
collision rate is low (<0.1%).
"""
def spend_utxo(self, tx_hash, tx_idx):
"""Spend a UTXO and return the 33-byte value.
If the UTXO is not in the cache it must be on disk. We store
all UTXOs so not finding one indicates a logic error or DB
corruption.
"""
def spend_utxo(self, tx_hash: bytes, nout: int):
# Fast track is it being in the cache # Fast track is it being in the cache
idx_packed = pack('<H', tx_idx) cache_value = self.utxo_cache.pop((tx_hash, nout), None)
cache_value = self.utxo_cache.pop(tx_hash + idx_packed, None)
if cache_value: if cache_value:
return cache_value return cache_value
# Spend it from the DB. prefix = Prefixes.hashX_utxo.pack_partial_key(tx_hash[:4])
# Key: b'h' + compressed_tx_hash + tx_idx + tx_num
# Value: hashX
prefix = DB_PREFIXES.HASHX_UTXO_PREFIX.value + tx_hash[:4] + idx_packed
candidates = {db_key: hashX for db_key, hashX in self.db.db.iterator(prefix=prefix)} candidates = {db_key: hashX for db_key, hashX in self.db.db.iterator(prefix=prefix)}
for hdb_key, hashX in candidates.items(): for hdb_key, hashX in candidates.items():
tx_num_packed = hdb_key[-4:] key = Prefixes.hashX_utxo.unpack_key(hdb_key)
if len(candidates) > 1: if len(candidates) > 1:
tx_num, = unpack('<I', tx_num_packed) hash = self.db.total_transactions[key.tx_num]
try:
hash, height = self.db.fs_tx_hash(tx_num)
except IndexError:
self.logger.error("data integrity error for hashx history: %s missing tx #%s (%s:%s)",
hashX.hex(), tx_num, hash_to_hex_str(tx_hash), tx_idx)
continue
if hash != tx_hash: if hash != tx_hash:
assert hash is not None # Should always be found assert hash is not None # Should always be found
continue continue
if key.nout != nout:
# Key: b'u' + address_hashX + tx_idx + tx_num continue
# Value: the UTXO value as a 64-bit unsigned integer udb_key = Prefixes.utxo.pack_key(hashX, key.tx_num, nout)
udb_key = DB_PREFIXES.UTXO_PREFIX.value + hashX + hdb_key[-6:]
utxo_value_packed = self.db.db.get(udb_key) utxo_value_packed = self.db.db.get(udb_key)
if utxo_value_packed is None: if utxo_value_packed is None:
self.logger.warning( self.logger.warning(
"%s:%s is not found in UTXO db for %s", hash_to_hex_str(tx_hash), tx_idx, hash_to_hex_str(hashX) "%s:%s is not found in UTXO db for %s", hash_to_hex_str(tx_hash), nout, hash_to_hex_str(hashX)
) )
raise ChainError(f"{hash_to_hex_str(tx_hash)}:{tx_idx} is not found in UTXO db for {hash_to_hex_str(hashX)}") raise ChainError(f"{hash_to_hex_str(tx_hash)}:{nout} is not found in UTXO db for {hash_to_hex_str(hashX)}")
# Remove both entries for this UTXO # Remove both entries for this UTXO
self.db_deletes.append(hdb_key) self.db_op_stack.extend([
self.db_deletes.append(udb_key) RevertableDelete(hdb_key, hashX),
RevertableDelete(udb_key, utxo_value_packed)
return hashX + tx_num_packed + utxo_value_packed ])
return hashX
self.logger.error('UTXO {hash_to_hex_str(tx_hash)} / {tx_idx} not found in "h" table') self.logger.error('UTXO {hash_to_hex_str(tx_hash)} / {tx_idx} not found in "h" table')
raise ChainError('UTXO {} / {:,d} not found in "h" table' raise ChainError('UTXO {} / {:,d} not found in "h" table'
.format(hash_to_hex_str(tx_hash), tx_idx)) .format(hash_to_hex_str(tx_hash), nout))
async def _process_prefetched_blocks(self): async def _process_prefetched_blocks(self):
"""Loop forever processing blocks as they arrive.""" """Loop forever processing blocks as they arrive."""
@ -1467,10 +1303,6 @@ class BlockProcessor:
self._caught_up_event.set() self._caught_up_event.set()
await self.blocks_event.wait() await self.blocks_event.wait()
self.blocks_event.clear() self.blocks_event.clear()
if self.reorg_count: # this could only happen by calling the reorg rpc
await self.reorg_chain(self.reorg_count)
self.reorg_count = 0
else:
blocks = self.prefetcher.get_prefetched_blocks() blocks = self.prefetcher.get_prefetched_blocks()
try: try:
await self.check_and_advance_blocks(blocks) await self.check_and_advance_blocks(blocks)
@ -1483,7 +1315,7 @@ class BlockProcessor:
# Flush everything but with first_sync->False state. # Flush everything but with first_sync->False state.
first_sync = self.db.first_sync first_sync = self.db.first_sync
self.db.first_sync = False self.db.first_sync = False
await self.flush() await self.write_state()
if first_sync: if first_sync:
self.logger.info(f'{lbry.__version__} synced to ' self.logger.info(f'{lbry.__version__} synced to '
f'height {self.height:,d}, halting here.') f'height {self.height:,d}, halting here.')

View file

@ -65,16 +65,10 @@ TXO_STRUCT_pack = TXO_STRUCT.pack
class FlushData: class FlushData:
height = attr.ib() height = attr.ib()
tx_count = attr.ib() tx_count = attr.ib()
headers = attr.ib()
block_hashes = attr.ib() block_hashes = attr.ib()
block_txs = attr.ib() block_txs = attr.ib()
claimtrie_stash = attr.ib() put_and_delete_ops = attr.ib()
# The following are flushed to the UTXO DB if undo_infos is not None
undo_infos = attr.ib()
adds = attr.ib()
deletes = attr.ib()
tip = attr.ib() tip = attr.ib()
undo = attr.ib()
OptionalResolveResultOrError = Optional[typing.Union[ResolveResult, LookupError, ValueError]] OptionalResolveResultOrError = Optional[typing.Union[ResolveResult, LookupError, ValueError]]
@ -143,9 +137,6 @@ class LevelDB:
self.merkle = Merkle() self.merkle = Merkle()
self.header_mc = MerkleCache(self.merkle, self.fs_block_hashes) self.header_mc = MerkleCache(self.merkle, self.fs_block_hashes)
self.headers_db = None
self.tx_db = None
self._tx_and_merkle_cache = LRUCacheWithMetrics(2 ** 17, metric_name='tx_and_merkle', namespace="wallet_server") self._tx_and_merkle_cache = LRUCacheWithMetrics(2 ** 17, metric_name='tx_and_merkle', namespace="wallet_server")
self.total_transactions = None self.total_transactions = None
self.transaction_num_mapping = {} self.transaction_num_mapping = {}
@ -748,61 +739,8 @@ class LevelDB:
raise RuntimeError(msg) raise RuntimeError(msg)
self.logger.info(f'flush count: {self.hist_flush_count:,d}') self.logger.info(f'flush count: {self.hist_flush_count:,d}')
# self.history.clear_excess(self.utxo_flush_count)
# < might happen at end of compaction as both DBs cannot be
# updated atomically
if self.hist_flush_count > self.utxo_flush_count:
self.logger.info('DB shut down uncleanly. Scanning for excess history flushes...')
keys = []
for key, hist in self.db.iterator(prefix=DB_PREFIXES.HASHX_HISTORY_PREFIX.value):
k = key[1:]
flush_id = int.from_bytes(k[-4:], byteorder='big')
if flush_id > self.utxo_flush_count:
keys.append(k)
self.logger.info(f'deleting {len(keys):,d} history entries')
self.hist_flush_count = self.utxo_flush_count
with self.db.write_batch() as batch:
for key in keys:
batch.delete(DB_PREFIXES.HASHX_HISTORY_PREFIX.value + key)
if keys:
self.logger.info('deleted %i excess history entries', len(keys))
self.utxo_flush_count = self.hist_flush_count self.utxo_flush_count = self.hist_flush_count
min_height = self.min_undo_height(self.db_height)
keys = []
for key, hist in self.db.iterator(prefix=DB_PREFIXES.UNDO_PREFIX.value):
height, = unpack('>I', key[-4:])
if height >= min_height:
break
keys.append(key)
if min_height > 0:
for key in self.db.iterator(start=DB_PREFIXES.undo_claimtrie.value,
stop=Prefixes.undo.pack_key(min_height),
include_value=False):
keys.append(key)
if keys:
with self.db.write_batch() as batch:
for key in keys:
batch.delete(key)
self.logger.info(f'deleted {len(keys):,d} stale undo entries')
# delete old block files
prefix = self.raw_block_prefix()
paths = [path for path in glob(f'{prefix}[0-9]*')
if len(path) > len(prefix)
and int(path[len(prefix):]) < min_height]
if paths:
for path in paths:
try:
os.remove(path)
except FileNotFoundError:
pass
self.logger.info(f'deleted {len(paths):,d} stale block files')
# Read TX counts (requires meta directory) # Read TX counts (requires meta directory)
await self._read_tx_counts() await self._read_tx_counts()
if self.total_transactions is None: if self.total_transactions is None:
@ -836,129 +774,50 @@ class LevelDB:
assert flush_data.tx_count == self.fs_tx_count == self.db_tx_count assert flush_data.tx_count == self.fs_tx_count == self.db_tx_count
assert flush_data.height == self.fs_height == self.db_height assert flush_data.height == self.fs_height == self.db_height
assert flush_data.tip == self.db_tip assert flush_data.tip == self.db_tip
assert not flush_data.headers
assert not flush_data.block_txs assert not flush_data.block_txs
assert not flush_data.adds assert not len(flush_data.put_and_delete_ops)
assert not flush_data.deletes
assert not flush_data.undo_infos
assert not self.hist_unflushed
def flush_dbs(self, flush_data: FlushData): def flush_dbs(self, flush_data: FlushData):
"""Flush out cached state. History is always flushed; UTXOs are
flushed if flush_utxos."""
if flush_data.height == self.db_height: if flush_data.height == self.db_height:
self.assert_flushed(flush_data) self.assert_flushed(flush_data)
return return
# start_time = time.time() min_height = self.min_undo_height(self.db_height)
prior_flush = self.last_flush delete_undo_keys = []
tx_delta = flush_data.tx_count - self.last_flush_tx_count if min_height > 0:
delete_undo_keys.extend(
# Flush to file system self.db.iterator(
# self.flush_fs(flush_data) start=Prefixes.undo.pack_key(0), stop=Prefixes.undo.pack_key(min_height), include_value=False
prior_tx_count = (self.tx_counts[self.fs_height] )
if self.fs_height >= 0 else 0) )
assert len(flush_data.block_txs) == len(flush_data.headers)
assert flush_data.height == self.fs_height + len(flush_data.headers)
assert flush_data.tx_count == (self.tx_counts[-1] if self.tx_counts
else 0)
assert len(self.tx_counts) == flush_data.height + 1
assert len(
b''.join(hashes for hashes, _ in flush_data.block_txs)
) // 32 == flush_data.tx_count - prior_tx_count, f"{len(b''.join(hashes for hashes, _ in flush_data.block_txs)) // 32} != {flush_data.tx_count}"
# Write the headers
# start_time = time.perf_counter()
with self.db.write_batch() as batch: with self.db.write_batch() as batch:
self.put = batch.put batch_put = batch.put
batch_put = self.put
batch_delete = batch.delete batch_delete = batch.delete
height_start = self.fs_height + 1
tx_num = prior_tx_count for staged_change in flush_data.put_and_delete_ops:
for i, (header, block_hash, (tx_hashes, txs)) in enumerate(
zip(flush_data.headers, flush_data.block_hashes, flush_data.block_txs)):
batch_put(DB_PREFIXES.HEADER_PREFIX.value + util.pack_be_uint64(height_start), header)
self.headers.append(header)
tx_count = self.tx_counts[height_start]
batch_put(DB_PREFIXES.BLOCK_HASH_PREFIX.value + util.pack_be_uint64(height_start), block_hash[::-1])
batch_put(DB_PREFIXES.TX_COUNT_PREFIX.value + util.pack_be_uint64(height_start), util.pack_be_uint64(tx_count))
height_start += 1
offset = 0
while offset < len(tx_hashes):
batch_put(DB_PREFIXES.TX_HASH_PREFIX.value + util.pack_be_uint64(tx_num), tx_hashes[offset:offset + 32])
batch_put(DB_PREFIXES.TX_NUM_PREFIX.value + tx_hashes[offset:offset + 32], util.pack_be_uint64(tx_num))
batch_put(DB_PREFIXES.TX_PREFIX.value + tx_hashes[offset:offset + 32], txs[offset // 32])
tx_num += 1
offset += 32
flush_data.headers.clear()
flush_data.block_txs.clear()
flush_data.block_hashes.clear()
for staged_change in flush_data.claimtrie_stash:
# print("ADVANCE", staged_change)
if staged_change.is_put: if staged_change.is_put:
batch_put(staged_change.key, staged_change.value) batch_put(staged_change.key, staged_change.value)
else: else:
batch_delete(staged_change.key) batch_delete(staged_change.key)
flush_data.claimtrie_stash.clear() for delete_key in delete_undo_keys:
batch_delete(delete_key)
for undo_ops, height in flush_data.undo:
batch_put(*Prefixes.undo.pack_item(height, undo_ops))
flush_data.undo.clear()
self.fs_height = flush_data.height self.fs_height = flush_data.height
self.fs_tx_count = flush_data.tx_count self.fs_tx_count = flush_data.tx_count
# Then history
self.hist_flush_count += 1 self.hist_flush_count += 1
flush_id = util.pack_be_uint32(self.hist_flush_count)
unflushed = self.hist_unflushed
for hashX in sorted(unflushed):
key = hashX + flush_id
batch_put(DB_PREFIXES.HASHX_HISTORY_PREFIX.value + key, unflushed[hashX].tobytes())
unflushed.clear()
self.hist_unflushed_count = 0 self.hist_unflushed_count = 0
#########################
# New undo information
for undo_info, height in flush_data.undo_infos:
batch_put(self.undo_key(height), b''.join(undo_info))
flush_data.undo_infos.clear()
# Spends
for key in sorted(flush_data.deletes):
batch_delete(key)
flush_data.deletes.clear()
# New UTXOs
for key, value in flush_data.adds.items():
# suffix = tx_idx + tx_num
hashX = value[:-12]
suffix = key[-2:] + value[-12:-8]
batch_put(DB_PREFIXES.HASHX_UTXO_PREFIX.value + key[:4] + suffix, hashX)
batch_put(DB_PREFIXES.UTXO_PREFIX.value + hashX + suffix, value[-8:])
flush_data.adds.clear()
self.utxo_flush_count = self.hist_flush_count self.utxo_flush_count = self.hist_flush_count
self.db_height = flush_data.height self.db_height = flush_data.height
self.db_tx_count = flush_data.tx_count self.db_tx_count = flush_data.tx_count
self.db_tip = flush_data.tip self.db_tip = flush_data.tip
self.last_flush_tx_count = self.fs_tx_count
now = time.time() now = time.time()
self.wall_time += now - self.last_flush self.wall_time += now - self.last_flush
self.last_flush = now self.last_flush = now
self.last_flush_tx_count = self.fs_tx_count
self.write_db_state(batch) self.write_db_state(batch)
def flush_backup(self, flush_data, touched): def flush_backup(self, flush_data):
"""Like flush_dbs() but when backing up. All UTXOs are flushed."""
assert not flush_data.headers
assert not flush_data.block_txs assert not flush_data.block_txs
assert flush_data.height < self.db_height assert flush_data.height < self.db_height
assert not self.hist_unflushed assert not self.hist_unflushed
@ -974,82 +833,25 @@ class LevelDB:
self.hist_flush_count += 1 self.hist_flush_count += 1
nremoves = 0 nremoves = 0
undo_ops = RevertableOpStack(self.db.get)
for (packed_ops, height) in reversed(flush_data.undo):
undo_ops.extend(reversed(RevertableOp.unpack_stack(packed_ops)))
undo_ops.append(
RevertableDelete(*Prefixes.undo.pack_item(height, packed_ops))
)
with self.db.write_batch() as batch: with self.db.write_batch() as batch:
batch_put = batch.put batch_put = batch.put
batch_delete = batch.delete batch_delete = batch.delete
for op in flush_data.put_and_delete_ops:
# print("flush undos", flush_data.undo_claimtrie)
for op in undo_ops:
# print("REWIND", op) # print("REWIND", op)
if op.is_put: if op.is_put:
batch_put(op.key, op.value) batch_put(op.key, op.value)
else: else:
batch_delete(op.key) batch_delete(op.key)
flush_data.undo.clear()
while self.fs_height > flush_data.height: while self.fs_height > flush_data.height:
self.fs_height -= 1 self.fs_height -= 1
self.headers.pop()
tx_count = flush_data.tx_count
for hashX in sorted(touched):
deletes = []
puts = {}
for key, hist in self.db.iterator(prefix=DB_PREFIXES.HASHX_HISTORY_PREFIX.value + hashX, reverse=True):
k = key[1:]
a = array.array('I')
a.frombytes(hist)
# Remove all history entries >= tx_count
idx = bisect_left(a, tx_count)
nremoves += len(a) - idx
if idx > 0:
puts[k] = a[:idx].tobytes()
break
deletes.append(k)
for key in deletes:
batch_delete(key)
for key, value in puts.items():
batch_put(key, value)
# New undo information
for undo_info, height in flush_data.undo:
batch.put(self.undo_key(height), b''.join(undo_info))
flush_data.undo.clear()
# Spends
for key in sorted(flush_data.deletes):
batch_delete(key)
flush_data.deletes.clear()
# New UTXOs
for key, value in flush_data.adds.items():
# suffix = tx_idx + tx_num
hashX = value[:-12]
suffix = key[-2:] + value[-12:-8]
batch_put(DB_PREFIXES.HASHX_UTXO_PREFIX.value + key[:4] + suffix, hashX)
batch_put(DB_PREFIXES.UTXO_PREFIX.value + hashX + suffix, value[-8:])
flush_data.adds.clear()
start_time = time.time() start_time = time.time()
add_count = len(flush_data.adds)
spend_count = len(flush_data.deletes) // 2
if self.db.for_sync: if self.db.for_sync:
block_count = flush_data.height - self.db_height block_count = flush_data.height - self.db_height
tx_count = flush_data.tx_count - self.db_tx_count tx_count = flush_data.tx_count - self.db_tx_count
elapsed = time.time() - start_time elapsed = time.time() - start_time
self.logger.info(f'flushed {block_count:,d} blocks with ' self.logger.info(f'flushed {block_count:,d} blocks with '
f'{tx_count:,d} txs, {add_count:,d} UTXO adds, ' f'{tx_count:,d} txs in '
f'{spend_count:,d} spends in '
f'{elapsed:.1f}s, committing...') f'{elapsed:.1f}s, committing...')
self.utxo_flush_count = self.hist_flush_count self.utxo_flush_count = self.hist_flush_count
@ -1121,7 +923,6 @@ class LevelDB:
return None, tx_height return None, tx_height
def _fs_transactions(self, txids: Iterable[str]): def _fs_transactions(self, txids: Iterable[str]):
unpack_be_uint64 = util.unpack_be_uint64
tx_counts = self.tx_counts tx_counts = self.tx_counts
tx_db_get = self.db.get tx_db_get = self.db.get
tx_cache = self._tx_and_merkle_cache tx_cache = self._tx_and_merkle_cache
@ -1133,14 +934,12 @@ class LevelDB:
tx, merkle = cached_tx tx, merkle = cached_tx
else: else:
tx_hash_bytes = bytes.fromhex(tx_hash)[::-1] tx_hash_bytes = bytes.fromhex(tx_hash)[::-1]
tx_num = tx_db_get(DB_PREFIXES.TX_NUM_PREFIX.value + tx_hash_bytes) tx_num = self.transaction_num_mapping.get(tx_hash_bytes)
tx = None tx = None
tx_height = -1 tx_height = -1
if tx_num is not None: if tx_num is not None:
tx_num = unpack_be_uint64(tx_num)
tx_height = bisect_right(tx_counts, tx_num) tx_height = bisect_right(tx_counts, tx_num)
if tx_height < self.db_height: tx = tx_db_get(Prefixes.tx.pack_key(tx_hash_bytes))
tx = tx_db_get(DB_PREFIXES.TX_PREFIX.value + tx_hash_bytes)
if tx_height == -1: if tx_height == -1:
merkle = { merkle = {
'block_height': -1 'block_height': -1
@ -1204,67 +1003,10 @@ class LevelDB:
def undo_key(self, height: int) -> bytes: def undo_key(self, height: int) -> bytes:
"""DB key for undo information at the given height.""" """DB key for undo information at the given height."""
return DB_PREFIXES.UNDO_PREFIX.value + pack('>I', height) return Prefixes.undo.pack_key(height)
def read_undo_info(self, height): def read_undo_info(self, height: int):
"""Read undo information from a file for the current height.""" return self.db.get(Prefixes.undo.pack_key(height))
return self.db.get(self.undo_key(height)), self.db.get(Prefixes.undo.pack_key(self.fs_height))
def raw_block_prefix(self):
return 'block'
def raw_block_path(self, height):
return os.path.join(self.env.db_dir, f'{self.raw_block_prefix()}{height:d}')
async def read_raw_block(self, height):
"""Returns a raw block read from disk. Raises FileNotFoundError
if the block isn't on-disk."""
def read():
with util.open_file(self.raw_block_path(height)) as f:
return f.read(-1)
return await asyncio.get_event_loop().run_in_executor(self.executor, read)
def write_raw_block(self, block, height):
"""Write a raw block to disk."""
with util.open_truncate(self.raw_block_path(height)) as f:
f.write(block)
# Delete old blocks to prevent them accumulating
try:
del_height = self.min_undo_height(height) - 1
os.remove(self.raw_block_path(del_height))
except FileNotFoundError:
pass
def clear_excess_undo_info(self):
"""Clear excess undo info. Only most recent N are kept."""
min_height = self.min_undo_height(self.db_height)
keys = []
for key, hist in self.db.iterator(prefix=DB_PREFIXES.UNDO_PREFIX.value):
height, = unpack('>I', key[-4:])
if height >= min_height:
break
keys.append(key)
if keys:
with self.db.write_batch() as batch:
for key in keys:
batch.delete(key)
self.logger.info(f'deleted {len(keys):,d} stale undo entries')
# delete old block files
prefix = self.raw_block_prefix()
paths = [path for path in glob(f'{prefix}[0-9]*')
if len(path) > len(prefix)
and int(path[len(prefix):]) < min_height]
if paths:
for path in paths:
try:
os.remove(path)
except FileNotFoundError:
pass
self.logger.info(f'deleted {len(paths):,d} stale block files')
# -- UTXO database # -- UTXO database

View file

@ -82,7 +82,7 @@ class BaseResolveTestCase(CommandTestCase):
check_supports(c['claimId'], c['supports']) check_supports(c['claimId'], c['supports'])
claim_hash = bytes.fromhex(c['claimId']) claim_hash = bytes.fromhex(c['claimId'])
self.assertEqual(c['validAtHeight'], db.get_activation( self.assertEqual(c['validAtHeight'], db.get_activation(
db.total_transactions.index(bytes.fromhex(c['txId'])[::-1]), c['n'] db.transaction_num_mapping[bytes.fromhex(c['txId'])[::-1]], c['n']
)) ))
self.assertEqual(c['effectiveAmount'], db.get_effective_amount(claim_hash)) self.assertEqual(c['effectiveAmount'], db.get_effective_amount(claim_hash))