fix(provider): move consistency check methods from NippyJarWriter to NippyJarChecker (#10633)

This commit is contained in:
joshieDo
2024-09-02 17:17:08 +01:00
committed by GitHub
parent 2c0170a03b
commit d58cf53850
7 changed files with 284 additions and 223 deletions

View File

@@ -62,9 +62,9 @@ mod tests {
StaticFileSegment, B256, U256,
};
use reth_provider::{
providers::StaticFileWriter, AccountExtReader, BlockReader, DatabaseProviderFactory,
ProviderFactory, ProviderResult, ReceiptProvider, StageCheckpointWriter,
StaticFileProviderFactory, StorageReader,
providers::{StaticFileProvider, StaticFileWriter},
AccountExtReader, BlockReader, DatabaseProviderFactory, ProviderFactory, ProviderResult,
ReceiptProvider, StageCheckpointWriter, StaticFileProviderFactory, StorageReader,
};
use reth_prune_types::{PruneMode, PruneModes};
use reth_stages_api::{
@@ -297,7 +297,10 @@ mod tests {
is_full_node: bool,
expected: Option<PipelineTarget>,
) {
let static_file_provider = db.factory.static_file_provider();
// We recreate the static file provider, since consistency heals are done on fetching the
// writer for the first time.
let static_file_provider =
StaticFileProvider::read_write(db.factory.static_file_provider().path()).unwrap();
// Simulate corruption by removing `prune_count` rows from the data file without updating
// its offset list and configuration.
@@ -312,8 +315,11 @@ mod tests {
data_file.get_ref().sync_all().unwrap();
}
// We recreate the static file provider, since consistency heals are done on fetching the
// writer for the first time.
assert_eq!(
static_file_provider
StaticFileProvider::read_write(db.factory.static_file_provider().path())
.unwrap()
.check_consistency(&db.factory.database_provider_ro().unwrap(), is_full_node,),
Ok(expected)
);

View File

@@ -0,0 +1,187 @@
use crate::{writer::OFFSET_SIZE_BYTES, NippyJar, NippyJarError, NippyJarHeader};
use std::{
cmp::Ordering,
fs::{File, OpenOptions},
io::{BufWriter, Seek, SeekFrom},
path::Path,
};
/// Performs consistency checks or heals on the [`NippyJar`] file
/// * Is the offsets file size expected?
/// * Is the data file size expected?
///
/// This is based on the assumption that [`NippyJar`] configuration is **always** the last one
/// to be updated when something is written, as by the `NippyJarWriter::commit()` function shows.
///
/// **For checks (read-only) use `check_consistency` method.**
///
/// **For heals (read-write) use `ensure_consistency` method.**
#[derive(Debug)]
pub struct NippyJarChecker<H: NippyJarHeader = ()> {
/// Associated [`NippyJar`], containing all necessary configurations for data
/// handling.
pub(crate) jar: NippyJar<H>,
/// File handle to where the data is stored.
pub(crate) data_file: Option<BufWriter<File>>,
/// File handle to where the offsets are stored.
pub(crate) offsets_file: Option<BufWriter<File>>,
}
impl<H: NippyJarHeader> NippyJarChecker<H> {
pub const fn new(jar: NippyJar<H>) -> Self {
Self { jar, data_file: None, offsets_file: None }
}
/// It will throw an error if the [`NippyJar`] is in a inconsistent state.
pub fn check_consistency(&mut self) -> Result<(), NippyJarError> {
self.handle_consistency(ConsistencyFailStrategy::ThrowError)
}
/// It will attempt to heal if the [`NippyJar`] is in a inconsistent state.
///
/// **ATTENTION**: disk commit should be handled externally by consuming `Self`
pub fn ensure_consistency(&mut self) -> Result<(), NippyJarError> {
self.handle_consistency(ConsistencyFailStrategy::Heal)
}
fn handle_consistency(&mut self, mode: ConsistencyFailStrategy) -> Result<(), NippyJarError> {
self.load_files(mode)?;
let reader = self.jar.open_data_reader()?;
// When an offset size is smaller than the initial (8), we are dealing with immutable
// data.
if reader.offset_size() != OFFSET_SIZE_BYTES {
return Err(NippyJarError::FrozenJar)
}
let expected_offsets_file_size: u64 = (1 + // first byte is the size of one offset
OFFSET_SIZE_BYTES as usize* self.jar.rows * self.jar.columns + // `offset size * num rows * num columns`
OFFSET_SIZE_BYTES as usize) as u64; // expected size of the data file
let actual_offsets_file_size = self.offsets_file().get_ref().metadata()?.len();
if mode.should_err() &&
expected_offsets_file_size.cmp(&actual_offsets_file_size) != Ordering::Equal
{
return Err(NippyJarError::InconsistentState)
}
// Offsets configuration wasn't properly committed
match expected_offsets_file_size.cmp(&actual_offsets_file_size) {
Ordering::Less => {
// Happened during an appending job
// TODO: ideally we could truncate until the last offset of the last column of the
// last row inserted
self.offsets_file().get_mut().set_len(expected_offsets_file_size)?;
}
Ordering::Greater => {
// Happened during a pruning job
// `num rows = (file size - 1 - size of one offset) / num columns`
self.jar.rows = ((actual_offsets_file_size.
saturating_sub(1). // first byte is the size of one offset
saturating_sub(OFFSET_SIZE_BYTES as u64) / // expected size of the data file
(self.jar.columns as u64)) /
OFFSET_SIZE_BYTES as u64) as usize;
// Freeze row count changed
self.jar.freeze_config()?;
}
Ordering::Equal => {}
}
// last offset should match the data_file_len
let last_offset = reader.reverse_offset(0)?;
let data_file_len = self.data_file().get_ref().metadata()?.len();
if mode.should_err() && last_offset.cmp(&data_file_len) != Ordering::Equal {
return Err(NippyJarError::InconsistentState)
}
// Offset list wasn't properly committed
match last_offset.cmp(&data_file_len) {
Ordering::Less => {
// Happened during an appending job, so we need to truncate the data, since there's
// no way to recover it.
self.data_file().get_mut().set_len(last_offset)?;
}
Ordering::Greater => {
// Happened during a pruning job, so we need to reverse iterate offsets until we
// find the matching one.
for index in 0..reader.offsets_count()? {
let offset = reader.reverse_offset(index + 1)?;
// It would only be equal if the previous row was fully pruned.
if offset <= data_file_len {
let new_len = self
.offsets_file()
.get_ref()
.metadata()?
.len()
.saturating_sub(OFFSET_SIZE_BYTES as u64 * (index as u64 + 1));
self.offsets_file().get_mut().set_len(new_len)?;
drop(reader);
// Since we decrease the offset list, we need to check the consistency of
// `self.jar.rows` again
self.handle_consistency(ConsistencyFailStrategy::Heal)?;
break
}
}
}
Ordering::Equal => {}
}
self.offsets_file().seek(SeekFrom::End(0))?;
self.data_file().seek(SeekFrom::End(0))?;
Ok(())
}
/// Loads data and offsets files.
fn load_files(&mut self, mode: ConsistencyFailStrategy) -> Result<(), NippyJarError> {
let load_file = |path: &Path| -> Result<BufWriter<File>, NippyJarError> {
let path = path
.exists()
.then_some(path)
.ok_or_else(|| NippyJarError::MissingFile(path.to_path_buf()))?;
Ok(BufWriter::new(OpenOptions::new().read(true).write(mode.should_heal()).open(path)?))
};
self.data_file = Some(load_file(self.jar.data_path())?);
self.offsets_file = Some(load_file(&self.jar.offsets_path())?);
Ok(())
}
/// Returns a mutable reference to offsets file.
///
/// **Panics** if it does not exist.
fn offsets_file(&mut self) -> &mut BufWriter<File> {
self.offsets_file.as_mut().expect("should exist")
}
/// Returns a mutable reference to data file.
///
/// **Panics** if it does not exist.
fn data_file(&mut self) -> &mut BufWriter<File> {
self.data_file.as_mut().expect("should exist")
}
}
/// Strategy on encountering an inconsistent state on [`NippyJarChecker`].
#[derive(Debug, Copy, Clone)]
enum ConsistencyFailStrategy {
/// Writer should heal.
Heal,
/// Writer should throw an error.
ThrowError,
}
impl ConsistencyFailStrategy {
/// Whether writer should heal.
const fn should_heal(&self) -> bool {
matches!(self, Self::Heal)
}
/// Whether writer should throw an error.
const fn should_err(&self) -> bool {
matches!(self, Self::ThrowError)
}
}

View File

@@ -1,3 +1,4 @@
use std::path::PathBuf;
use thiserror::Error;
/// Errors associated with [`crate::NippyJar`].
@@ -60,4 +61,6 @@ pub enum NippyJarError {
FrozenJar,
#[error("File is in an inconsistent state.")]
InconsistentState,
#[error("Missing file: {0}.")]
MissingFile(PathBuf),
}

View File

@@ -44,7 +44,10 @@ mod cursor;
pub use cursor::NippyJarCursor;
mod writer;
pub use writer::{ConsistencyFailStrategy, NippyJarWriter};
pub use writer::NippyJarWriter;
mod consistency;
pub use consistency::NippyJarChecker;
const NIPPY_JAR_VERSION: usize = 1;
@@ -346,7 +349,7 @@ impl<H: NippyJarHeader> NippyJar<H> {
self.freeze_filters()?;
// Creates the writer, data and offsets file
let mut writer = NippyJarWriter::new(self, ConsistencyFailStrategy::Heal)?;
let mut writer = NippyJarWriter::new(self)?;
// Append rows to file while holding offsets in memory
writer.append_rows(columns, total_rows)?;
@@ -959,7 +962,7 @@ mod tests {
assert!(initial_offset_size > 0);
// Appends a third row
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
writer.append_column(Some(Ok(&col1[2]))).unwrap();
writer.append_column(Some(Ok(&col2[2]))).unwrap();
@@ -990,7 +993,7 @@ mod tests {
// Writer will execute a consistency check and verify first that the offset list on disk
// doesn't match the nippy.rows, and prune it. Then, it will prune the data file
// accordingly as well.
let writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let writer = NippyJarWriter::new(nippy).unwrap();
assert_eq!(initial_rows, writer.rows());
assert_eq!(
initial_offset_size,
@@ -1016,7 +1019,7 @@ mod tests {
// Appends a third row, so we have an offset list in memory, which is not flushed to disk,
// while the data has been.
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
writer.append_column(Some(Ok(&col1[2]))).unwrap();
writer.append_column(Some(Ok(&col2[2]))).unwrap();
@@ -1039,7 +1042,7 @@ mod tests {
// Writer will execute a consistency check and verify that the data file has more data than
// it should, and resets it to the last offset of the list (on disk here)
let writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let writer = NippyJarWriter::new(nippy).unwrap();
assert_eq!(initial_rows, writer.rows());
assert_eq!(
initial_data_size,
@@ -1055,7 +1058,7 @@ mod tests {
assert_eq!(nippy.max_row_size, 0);
assert_eq!(nippy.rows, 0);
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
assert_eq!(writer.column(), 0);
writer.append_column(Some(Ok(&col1[0]))).unwrap();
@@ -1093,7 +1096,7 @@ mod tests {
assert_eq!(nippy.max_row_size, col1[0].len() + col2[0].len());
assert_eq!(nippy.rows, 1);
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
assert_eq!(writer.column(), 0);
writer.append_column(Some(Ok(&col1[1]))).unwrap();
@@ -1124,7 +1127,7 @@ mod tests {
fn prune_rows(num_columns: usize, file_path: &Path, col1: &[Vec<u8>], col2: &[Vec<u8>]) {
let nippy = NippyJar::load_without_header(file_path).unwrap();
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
// Appends a third row, so we have an offset list in memory, which is not flushed to disk
writer.append_column(Some(Ok(&col1[2]))).unwrap();
@@ -1155,7 +1158,7 @@ mod tests {
}
// This should prune from the ondisk offset list and clear the jar.
let mut writer = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let mut writer = NippyJarWriter::new(nippy).unwrap();
writer.prune_rows(1).unwrap();
assert!(writer.is_dirty());
@@ -1196,6 +1199,6 @@ mod tests {
data_file.set_len(data_len - 32 * missing_offsets).unwrap();
// runs the consistency check.
let _ = NippyJarWriter::new(nippy, ConsistencyFailStrategy::Heal).unwrap();
let _ = NippyJarWriter::new(nippy).unwrap();
}
}

View File

@@ -1,13 +1,15 @@
use crate::{compression::Compression, ColumnResult, NippyJar, NippyJarError, NippyJarHeader};
use crate::{
compression::Compression, ColumnResult, NippyJar, NippyJarChecker, NippyJarError,
NippyJarHeader,
};
use std::{
cmp::Ordering,
fs::{File, OpenOptions},
io::{BufWriter, Read, Seek, SeekFrom, Write},
path::Path,
};
/// Size of one offset in bytes.
const OFFSET_SIZE_BYTES: u8 = 8;
pub(crate) const OFFSET_SIZE_BYTES: u8 = 8;
/// Writer of [`NippyJar`]. Handles table data and offsets only.
///
@@ -46,22 +48,32 @@ pub struct NippyJarWriter<H: NippyJarHeader = ()> {
impl<H: NippyJarHeader> NippyJarWriter<H> {
/// Creates a [`NippyJarWriter`] from [`NippyJar`].
///
/// If `read_only` is set to `true`, any inconsistency issue won't be healed, and will return
/// [`NippyJarError::InconsistentState`] instead.
pub fn new(
jar: NippyJar<H>,
check_mode: ConsistencyFailStrategy,
) -> Result<Self, NippyJarError> {
/// If will **always** attempt to heal any inconsistent state when called.
pub fn new(jar: NippyJar<H>) -> Result<Self, NippyJarError> {
let (data_file, offsets_file, is_created) =
Self::create_or_open_files(jar.data_path(), &jar.offsets_path())?;
// Makes sure we don't have dangling data and offset files
jar.freeze_config()?;
let (jar, data_file, offsets_file) = if is_created {
// Makes sure we don't have dangling data and offset files when we just created the file
jar.freeze_config()?;
(jar, BufWriter::new(data_file), BufWriter::new(offsets_file))
} else {
// If we are opening a previously created jar, we need to check its consistency, and
// make changes if necessary.
let mut checker = NippyJarChecker::new(jar);
checker.ensure_consistency()?;
let NippyJarChecker { jar, data_file, offsets_file } = checker;
// Calling ensure_consistency, will fill data_file and offsets_file
(jar, data_file.expect("qed"), offsets_file.expect("qed"))
};
let mut writer = Self {
jar,
data_file: BufWriter::new(data_file),
offsets_file: BufWriter::new(offsets_file),
data_file,
offsets_file,
tmp_buf: Vec::with_capacity(1_000_000),
uncompressed_row_size: 0,
offsets: Vec::with_capacity(1_000_000),
@@ -69,13 +81,9 @@ impl<H: NippyJarHeader> NippyJarWriter<H> {
dirty: false,
};
// If we are opening a previously created jar, we need to check its consistency, and make
// changes if necessary.
if !is_created {
writer.ensure_file_consistency(check_mode)?;
if check_mode.should_heal() {
writer.commit()?;
}
// Commit any potential heals done above.
writer.commit()?;
}
Ok(writer)
@@ -147,107 +155,6 @@ impl<H: NippyJarHeader> NippyJarWriter<H> {
Ok((data_file, offsets_file, is_created))
}
/// Performs consistency checks on the [`NippyJar`] file and might self-heal or throw an error
/// according to [`ConsistencyFailStrategy`].
/// * Is the offsets file size expected?
/// * Is the data file size expected?
///
/// This is based on the assumption that [`NippyJar`] configuration is **always** the last one
/// to be updated when something is written, as by the `commit()` function shows.
pub fn ensure_file_consistency(
&mut self,
check_mode: ConsistencyFailStrategy,
) -> Result<(), NippyJarError> {
let reader = self.jar.open_data_reader()?;
// When an offset size is smaller than the initial (8), we are dealing with immutable
// data.
if reader.offset_size() != OFFSET_SIZE_BYTES {
return Err(NippyJarError::FrozenJar)
}
let expected_offsets_file_size: u64 = (1 + // first byte is the size of one offset
OFFSET_SIZE_BYTES as usize* self.jar.rows * self.jar.columns + // `offset size * num rows * num columns`
OFFSET_SIZE_BYTES as usize) as u64; // expected size of the data file
let actual_offsets_file_size = self.offsets_file.get_ref().metadata()?.len();
if check_mode.should_err() &&
expected_offsets_file_size.cmp(&actual_offsets_file_size) != Ordering::Equal
{
return Err(NippyJarError::InconsistentState)
}
// Offsets configuration wasn't properly committed
match expected_offsets_file_size.cmp(&actual_offsets_file_size) {
Ordering::Less => {
// Happened during an appending job
// TODO: ideally we could truncate until the last offset of the last column of the
// last row inserted
self.offsets_file.get_mut().set_len(expected_offsets_file_size)?;
}
Ordering::Greater => {
// Happened during a pruning job
// `num rows = (file size - 1 - size of one offset) / num columns`
self.jar.rows = ((actual_offsets_file_size.
saturating_sub(1). // first byte is the size of one offset
saturating_sub(OFFSET_SIZE_BYTES as u64) / // expected size of the data file
(self.jar.columns as u64)) /
OFFSET_SIZE_BYTES as u64) as usize;
// Freeze row count changed
self.jar.freeze_config()?;
}
Ordering::Equal => {}
}
// last offset should match the data_file_len
let last_offset = reader.reverse_offset(0)?;
let data_file_len = self.data_file.get_ref().metadata()?.len();
if check_mode.should_err() && last_offset.cmp(&data_file_len) != Ordering::Equal {
return Err(NippyJarError::InconsistentState)
}
// Offset list wasn't properly committed
match last_offset.cmp(&data_file_len) {
Ordering::Less => {
// Happened during an appending job, so we need to truncate the data, since there's
// no way to recover it.
self.data_file.get_mut().set_len(last_offset)?;
}
Ordering::Greater => {
// Happened during a pruning job, so we need to reverse iterate offsets until we
// find the matching one.
for index in 0..reader.offsets_count()? {
let offset = reader.reverse_offset(index + 1)?;
// It would only be equal if the previous row was fully pruned.
if offset <= data_file_len {
let new_len = self
.offsets_file
.get_ref()
.metadata()?
.len()
.saturating_sub(OFFSET_SIZE_BYTES as u64 * (index as u64 + 1));
self.offsets_file.get_mut().set_len(new_len)?;
drop(reader);
// Since we decrease the offset list, we need to check the consistency of
// `self.jar.rows` again
self.ensure_file_consistency(ConsistencyFailStrategy::Heal)?;
break
}
}
}
Ordering::Equal => {}
}
self.offsets_file.seek(SeekFrom::End(0))?;
self.data_file.seek(SeekFrom::End(0))?;
Ok(())
}
/// Appends rows to data file. `fn commit()` should be called to flush offsets and config to
/// disk.
///
@@ -540,24 +447,3 @@ impl<H: NippyJarHeader> NippyJarWriter<H> {
&self.jar
}
}
/// Strategy on encountering an inconsistent state when creating a [`NippyJarWriter`].
#[derive(Debug, Copy, Clone)]
pub enum ConsistencyFailStrategy {
/// Writer should heal.
Heal,
/// Writer should throw an error.
ThrowError,
}
impl ConsistencyFailStrategy {
/// Whether writer should heal.
const fn should_heal(&self) -> bool {
matches!(self, Self::Heal)
}
/// Whether writer should throw an error.
const fn should_err(&self) -> bool {
matches!(self, Self::ThrowError)
}
}

View File

@@ -22,7 +22,7 @@ use reth_db_api::{
table::Table,
transaction::DbTx,
};
use reth_nippy_jar::NippyJar;
use reth_nippy_jar::{NippyJar, NippyJarChecker};
use reth_primitives::{
keccak256,
static_file::{find_fixed_range, HighestStaticFiles, SegmentHeader, SegmentRangeInclusive},
@@ -573,7 +573,12 @@ impl StaticFileProvider {
// * pruning data was interrupted before a config commit, then we have deleted data that
// we are expected to still have. We need to check the Database and unwind everything
// accordingly.
self.ensure_file_consistency(segment)?;
if self.access.is_read_only() {
self.check_segment_consistency(segment)?;
} else {
// Fetching the writer will attempt to heal any file level inconsistency.
self.latest_writer(segment)?;
}
// Only applies to block-based static files. (Headers)
//
@@ -655,6 +660,23 @@ impl StaticFileProvider {
Ok(unwind_target.map(PipelineTarget::Unwind))
}
/// Checks consistency of the latest static file segment and throws an error if at fault.
/// Read-only.
pub fn check_segment_consistency(&self, segment: StaticFileSegment) -> ProviderResult<()> {
if let Some(latest_block) = self.get_highest_static_file_block(segment) {
let file_path =
self.directory().join(segment.filename(&find_fixed_range(latest_block)));
let jar = NippyJar::<SegmentHeader>::load(&file_path)
.map_err(|e| ProviderError::NippyJar(e.to_string()))?;
NippyJarChecker::new(jar)
.check_consistency()
.map_err(|e| ProviderError::NippyJar(e.to_string()))?;
}
Ok(())
}
/// Check invariants for each corresponding table and static file segment:
///
/// * the corresponding database table should overlap or have continuity in their keys
@@ -1040,9 +1062,6 @@ pub trait StaticFileWriter {
/// Commits all changes of all [`StaticFileProviderRW`] of all [`StaticFileSegment`].
fn commit(&self) -> ProviderResult<()>;
/// Checks consistency of the segment latest file and heals if possible.
fn ensure_file_consistency(&self, segment: StaticFileSegment) -> ProviderResult<()>;
}
impl StaticFileWriter for StaticFileProvider {
@@ -1071,28 +1090,6 @@ impl StaticFileWriter for StaticFileProvider {
fn commit(&self) -> ProviderResult<()> {
self.writers.commit()
}
fn ensure_file_consistency(&self, segment: StaticFileSegment) -> ProviderResult<()> {
match self.access {
StaticFileAccess::RO => {
let latest_block = self.get_highest_static_file_block(segment).unwrap_or_default();
let mut writer = StaticFileProviderRW::new(
segment,
latest_block,
Arc::downgrade(&self.0),
self.metrics.clone(),
)?;
writer.ensure_file_consistency(self.access.is_read_only())?;
}
StaticFileAccess::RW => {
self.latest_writer(segment)?.ensure_file_consistency(self.access.is_read_only())?;
}
}
Ok(())
}
}
impl HeaderProvider for StaticFileProvider {

View File

@@ -5,7 +5,7 @@ use crate::providers::static_file::metrics::StaticFileProviderOperation;
use parking_lot::{lock_api::RwLockWriteGuard, RawRwLock, RwLock};
use reth_codecs::Compact;
use reth_db_api::models::CompactU256;
use reth_nippy_jar::{ConsistencyFailStrategy, NippyJar, NippyJarError, NippyJarWriter};
use reth_nippy_jar::{NippyJar, NippyJarError, NippyJarWriter};
use reth_primitives::{
static_file::{find_fixed_range, SegmentHeader, SegmentRangeInclusive},
BlockHash, BlockNumber, Header, Receipt, StaticFileSegment, TransactionSignedNoHash, TxNumber,
@@ -105,6 +105,9 @@ pub struct StaticFileProviderRW {
impl StaticFileProviderRW {
/// Creates a new [`StaticFileProviderRW`] for a [`StaticFileSegment`].
///
/// Before use, transaction based segments should ensure the block end range is the expected
/// one, and heal if not. For more check `Self::ensure_end_range_consistency`.
pub fn new(
segment: StaticFileSegment,
block: BlockNumber,
@@ -112,14 +115,18 @@ impl StaticFileProviderRW {
metrics: Option<Arc<StaticFileProviderMetrics>>,
) -> ProviderResult<Self> {
let (writer, data_path) = Self::open(segment, block, reader.clone(), metrics.clone())?;
Ok(Self {
let mut writer = Self {
writer,
data_path,
buf: Vec::with_capacity(100),
reader,
metrics,
prune_on_commit: None,
})
};
writer.ensure_end_range_consistency()?;
Ok(writer)
}
fn open(
@@ -150,14 +157,7 @@ impl StaticFileProviderRW {
Err(err) => return Err(err),
};
let reader = Self::upgrade_provider_to_strong_reference(&reader);
let access = if reader.is_read_only() {
ConsistencyFailStrategy::ThrowError
} else {
ConsistencyFailStrategy::Heal
};
let result = match NippyJarWriter::new(jar, access) {
let result = match NippyJarWriter::new(jar) {
Ok(writer) => Ok((writer, path)),
Err(NippyJarError::FrozenJar) => {
// This static file has been frozen, so we should
@@ -177,33 +177,15 @@ impl StaticFileProviderRW {
Ok(result)
}
/// Checks the consistency of the file and heals it if necessary and `read_only` is set to
/// false. If the check fails, it will return an error.
/// If a file level healing happens, we need to update the end range on the
/// [`SegmentHeader`].
///
/// If healing does happen, it will update the end range on the [`SegmentHeader`]. However, for
/// transaction based segments, the block end range has to be found and healed externally.
/// However, for transaction based segments, the block end range has to be found and healed
/// externally.
///
/// Check [`NippyJarWriter::ensure_file_consistency`] for more on healing.
pub fn ensure_file_consistency(&mut self, read_only: bool) -> ProviderResult<()> {
let inconsistent_error = || {
ProviderError::NippyJar(
"Inconsistent state found. Restart the node to heal.".to_string(),
)
};
let check_mode = if read_only {
ConsistencyFailStrategy::ThrowError
} else {
ConsistencyFailStrategy::Heal
};
self.writer.ensure_file_consistency(check_mode).map_err(|error| {
if matches!(error, NippyJarError::InconsistentState) {
return inconsistent_error()
}
ProviderError::NippyJar(error.to_string())
})?;
/// Check [`reth_nippy_jar::NippyJarChecker`] &
/// [`NippyJarWriter`] for more on healing.
fn ensure_end_range_consistency(&mut self) -> ProviderResult<()> {
// If we have lost rows (in this run or previous), we need to update the [SegmentHeader].
let expected_rows = if self.user_header().segment().is_headers() {
self.user_header().block_len().unwrap_or_default()
@@ -212,9 +194,6 @@ impl StaticFileProviderRW {
};
let pruned_rows = expected_rows - self.writer.rows() as u64;
if pruned_rows > 0 {
if read_only {
return Err(inconsistent_error())
}
self.user_header_mut().prune(pruned_rows);
}