[pbs-devel] [RFC v2 proxmox-backup 06/42] verify: refactor verify related functions to be methods of worker
Christian Ebner
c.ebner at proxmox.com
Thu May 29 16:31:31 CEST 2025
Instead of passing the VerifyWorker state as reference to the various
verification related functions, implement them as methods or
associated functions of the VerifyWorker. This does not only make
their correlation more clear, but it also reduces the number of
function call parameters and improves readability.
No functional changes intended.
Signed-off-by: Christian Ebner <c.ebner at proxmox.com>
---
src/api2/admin/datastore.rs | 28 +-
src/api2/backup/environment.rs | 7 +-
src/backup/verify.rs | 830 ++++++++++++++++-----------------
src/server/verify_job.rs | 12 +-
4 files changed, 423 insertions(+), 454 deletions(-)
diff --git a/src/api2/admin/datastore.rs b/src/api2/admin/datastore.rs
index 392494488..7dc881ade 100644
--- a/src/api2/admin/datastore.rs
+++ b/src/api2/admin/datastore.rs
@@ -70,10 +70,7 @@ use proxmox_rest_server::{formatter, WorkerTask};
use crate::api2::backup::optional_ns_param;
use crate::api2::node::rrd::create_value_from_rrd;
-use crate::backup::{
- check_ns_privs_full, verify_all_backups, verify_backup_dir, verify_backup_group, verify_filter,
- ListAccessibleBackupGroups, NS_PRIVS_OK,
-};
+use crate::backup::{check_ns_privs_full, ListAccessibleBackupGroups, VerifyWorker, NS_PRIVS_OK};
use crate::server::jobstate::{compute_schedule_status, Job, JobState};
@@ -896,14 +893,15 @@ pub fn verify(
auth_id.to_string(),
to_stdout,
move |worker| {
- let verify_worker = crate::backup::VerifyWorker::new(worker.clone(), datastore);
+ let verify_worker = VerifyWorker::new(worker.clone(), datastore);
let failed_dirs = if let Some(backup_dir) = backup_dir {
let mut res = Vec::new();
- if !verify_backup_dir(
- &verify_worker,
+ if !verify_worker.verify_backup_dir(
&backup_dir,
worker.upid().clone(),
- Some(&move |manifest| verify_filter(ignore_verified, outdated_after, manifest)),
+ Some(&move |manifest| {
+ VerifyWorker::verify_filter(ignore_verified, outdated_after, manifest)
+ }),
)? {
res.push(print_ns_and_snapshot(
backup_dir.backup_ns(),
@@ -912,12 +910,13 @@ pub fn verify(
}
res
} else if let Some(backup_group) = backup_group {
- verify_backup_group(
- &verify_worker,
+ verify_worker.verify_backup_group(
&backup_group,
&mut StoreProgress::new(1),
worker.upid(),
- Some(&move |manifest| verify_filter(ignore_verified, outdated_after, manifest)),
+ Some(&move |manifest| {
+ VerifyWorker::verify_filter(ignore_verified, outdated_after, manifest)
+ }),
)?
} else {
let owner = if owner_check_required {
@@ -926,13 +925,14 @@ pub fn verify(
None
};
- verify_all_backups(
- &verify_worker,
+ verify_worker.verify_all_backups(
worker.upid(),
ns,
max_depth,
owner,
- Some(&move |manifest| verify_filter(ignore_verified, outdated_after, manifest)),
+ Some(&move |manifest| {
+ VerifyWorker::verify_filter(ignore_verified, outdated_after, manifest)
+ }),
)?
};
if !failed_dirs.is_empty() {
diff --git a/src/api2/backup/environment.rs b/src/api2/backup/environment.rs
index 3d541b461..6cd29f512 100644
--- a/src/api2/backup/environment.rs
+++ b/src/api2/backup/environment.rs
@@ -18,7 +18,7 @@ use pbs_datastore::fixed_index::FixedIndexWriter;
use pbs_datastore::{DataBlob, DataStore};
use proxmox_rest_server::{formatter::*, WorkerTask};
-use crate::backup::verify_backup_dir_with_lock;
+use crate::backup::VerifyWorker;
use hyper::{Body, Response};
@@ -671,9 +671,8 @@ impl BackupEnvironment {
move |worker| {
worker.log_message("Automatically verifying newly added snapshot");
- let verify_worker = crate::backup::VerifyWorker::new(worker.clone(), datastore);
- if !verify_backup_dir_with_lock(
- &verify_worker,
+ let verify_worker = VerifyWorker::new(worker.clone(), datastore);
+ if !verify_worker.verify_backup_dir_with_lock(
&backup_dir,
worker.upid().clone(),
None,
diff --git a/src/backup/verify.rs b/src/backup/verify.rs
index 3d2cba8ac..0b954ae23 100644
--- a/src/backup/verify.rs
+++ b/src/backup/verify.rs
@@ -44,517 +44,491 @@ impl VerifyWorker {
corrupt_chunks: Arc::new(Mutex::new(HashSet::with_capacity(64))),
}
}
-}
-
-fn verify_blob(backup_dir: &BackupDir, info: &FileInfo) -> Result<(), Error> {
- let blob = backup_dir.load_blob(&info.filename)?;
- let raw_size = blob.raw_size();
- if raw_size != info.size {
- bail!("wrong size ({} != {})", info.size, raw_size);
- }
-
- let csum = openssl::sha::sha256(blob.raw_data());
- if csum != info.csum {
- bail!("wrong index checksum");
- }
+ fn verify_blob(backup_dir: &BackupDir, info: &FileInfo) -> Result<(), Error> {
+ let blob = backup_dir.load_blob(&info.filename)?;
- match blob.crypt_mode()? {
- CryptMode::Encrypt => Ok(()),
- CryptMode::None => {
- // digest already verified above
- blob.decode(None, None)?;
- Ok(())
+ let raw_size = blob.raw_size();
+ if raw_size != info.size {
+ bail!("wrong size ({} != {})", info.size, raw_size);
}
- CryptMode::SignOnly => bail!("Invalid CryptMode for blob"),
- }
-}
-
-fn rename_corrupted_chunk(datastore: Arc<DataStore>, digest: &[u8; 32]) {
- let (path, digest_str) = datastore.chunk_path(digest);
- let mut counter = 0;
- let mut new_path = path.clone();
- loop {
- new_path.set_file_name(format!("{}.{}.bad", digest_str, counter));
- if new_path.exists() && counter < 9 {
- counter += 1;
- } else {
- break;
+ let csum = openssl::sha::sha256(blob.raw_data());
+ if csum != info.csum {
+ bail!("wrong index checksum");
}
- }
- match std::fs::rename(&path, &new_path) {
- Ok(_) => {
- info!("corrupted chunk renamed to {:?}", &new_path);
- }
- Err(err) => {
- match err.kind() {
- std::io::ErrorKind::NotFound => { /* ignored */ }
- _ => info!("could not rename corrupted chunk {:?} - {err}", &path),
+ match blob.crypt_mode()? {
+ CryptMode::Encrypt => Ok(()),
+ CryptMode::None => {
+ // digest already verified above
+ blob.decode(None, None)?;
+ Ok(())
}
+ CryptMode::SignOnly => bail!("Invalid CryptMode for blob"),
}
- };
-}
+ }
-fn verify_index_chunks(
- verify_worker: &VerifyWorker,
- index: Box<dyn IndexFile + Send>,
- crypt_mode: CryptMode,
-) -> Result<(), Error> {
- let errors = Arc::new(AtomicUsize::new(0));
+ fn rename_corrupted_chunk(datastore: Arc<DataStore>, digest: &[u8; 32]) {
+ let (path, digest_str) = datastore.chunk_path(digest);
- let start_time = Instant::now();
+ let mut counter = 0;
+ let mut new_path = path.clone();
+ loop {
+ new_path.set_file_name(format!("{}.{}.bad", digest_str, counter));
+ if new_path.exists() && counter < 9 {
+ counter += 1;
+ } else {
+ break;
+ }
+ }
- let mut read_bytes = 0;
- let mut decoded_bytes = 0;
+ match std::fs::rename(&path, &new_path) {
+ Ok(_) => {
+ info!("corrupted chunk renamed to {:?}", &new_path);
+ }
+ Err(err) => {
+ match err.kind() {
+ std::io::ErrorKind::NotFound => { /* ignored */ }
+ _ => info!("could not rename corrupted chunk {:?} - {err}", &path),
+ }
+ }
+ };
+ }
- let datastore2 = Arc::clone(&verify_worker.datastore);
- let corrupt_chunks2 = Arc::clone(&verify_worker.corrupt_chunks);
- let verified_chunks2 = Arc::clone(&verify_worker.verified_chunks);
- let errors2 = Arc::clone(&errors);
+ fn verify_index_chunks(
+ &self,
+ index: Box<dyn IndexFile + Send>,
+ crypt_mode: CryptMode,
+ ) -> Result<(), Error> {
+ let errors = Arc::new(AtomicUsize::new(0));
+
+ let start_time = Instant::now();
+
+ let mut read_bytes = 0;
+ let mut decoded_bytes = 0;
+
+ let datastore2 = Arc::clone(&self.datastore);
+ let corrupt_chunks2 = Arc::clone(&self.corrupt_chunks);
+ let verified_chunks2 = Arc::clone(&self.verified_chunks);
+ let errors2 = Arc::clone(&errors);
+
+ let decoder_pool = ParallelHandler::new(
+ "verify chunk decoder",
+ 4,
+ move |(chunk, digest, size): (DataBlob, [u8; 32], u64)| {
+ let chunk_crypt_mode = match chunk.crypt_mode() {
+ Err(err) => {
+ corrupt_chunks2.lock().unwrap().insert(digest);
+ info!("can't verify chunk, unknown CryptMode - {err}");
+ errors2.fetch_add(1, Ordering::SeqCst);
+ return Ok(());
+ }
+ Ok(mode) => mode,
+ };
+
+ if chunk_crypt_mode != crypt_mode {
+ info!(
+ "chunk CryptMode {chunk_crypt_mode:?} does not match index CryptMode {crypt_mode:?}"
+ );
+ errors2.fetch_add(1, Ordering::SeqCst);
+ }
- let decoder_pool = ParallelHandler::new(
- "verify chunk decoder",
- 4,
- move |(chunk, digest, size): (DataBlob, [u8; 32], u64)| {
- let chunk_crypt_mode = match chunk.crypt_mode() {
- Err(err) => {
+ if let Err(err) = chunk.verify_unencrypted(size as usize, &digest) {
corrupt_chunks2.lock().unwrap().insert(digest);
- info!("can't verify chunk, unknown CryptMode - {err}");
+ info!("{err}");
errors2.fetch_add(1, Ordering::SeqCst);
- return Ok(());
+ Self::rename_corrupted_chunk(datastore2.clone(), &digest);
+ } else {
+ verified_chunks2.lock().unwrap().insert(digest);
}
- Ok(mode) => mode,
- };
- if chunk_crypt_mode != crypt_mode {
- info!(
- "chunk CryptMode {chunk_crypt_mode:?} does not match index CryptMode {crypt_mode:?}"
- );
- errors2.fetch_add(1, Ordering::SeqCst);
- }
+ Ok(())
+ },
+ );
- if let Err(err) = chunk.verify_unencrypted(size as usize, &digest) {
- corrupt_chunks2.lock().unwrap().insert(digest);
- info!("{err}");
- errors2.fetch_add(1, Ordering::SeqCst);
- rename_corrupted_chunk(datastore2.clone(), &digest);
+ let skip_chunk = |digest: &[u8; 32]| -> bool {
+ if self.verified_chunks.lock().unwrap().contains(digest) {
+ true
+ } else if self.corrupt_chunks.lock().unwrap().contains(digest) {
+ let digest_str = hex::encode(digest);
+ info!("chunk {digest_str} was marked as corrupt");
+ errors.fetch_add(1, Ordering::SeqCst);
+ true
} else {
- verified_chunks2.lock().unwrap().insert(digest);
+ false
}
+ };
+ let check_abort = |pos: usize| -> Result<(), Error> {
+ if pos & 1023 == 0 {
+ self.worker.check_abort()?;
+ self.worker.fail_on_shutdown()?;
+ }
Ok(())
- },
- );
-
- let skip_chunk = |digest: &[u8; 32]| -> bool {
- if verify_worker
- .verified_chunks
- .lock()
- .unwrap()
- .contains(digest)
- {
- true
- } else if verify_worker
- .corrupt_chunks
- .lock()
- .unwrap()
- .contains(digest)
- {
- let digest_str = hex::encode(digest);
- info!("chunk {digest_str} was marked as corrupt");
- errors.fetch_add(1, Ordering::SeqCst);
- true
- } else {
- false
- }
- };
-
- let check_abort = |pos: usize| -> Result<(), Error> {
- if pos & 1023 == 0 {
- verify_worker.worker.check_abort()?;
- verify_worker.worker.fail_on_shutdown()?;
- }
- Ok(())
- };
+ };
- let chunk_list =
- verify_worker
+ let chunk_list = self
.datastore
.get_chunks_in_order(&*index, skip_chunk, check_abort)?;
- for (pos, _) in chunk_list {
- verify_worker.worker.check_abort()?;
- verify_worker.worker.fail_on_shutdown()?;
+ for (pos, _) in chunk_list {
+ self.worker.check_abort()?;
+ self.worker.fail_on_shutdown()?;
- let info = index.chunk_info(pos).unwrap();
+ let info = index.chunk_info(pos).unwrap();
- // we must always recheck this here, the parallel worker below alter it!
- if skip_chunk(&info.digest) {
- continue; // already verified or marked corrupt
- }
-
- match verify_worker.datastore.load_chunk(&info.digest) {
- Err(err) => {
- verify_worker
- .corrupt_chunks
- .lock()
- .unwrap()
- .insert(info.digest);
- error!("can't verify chunk, load failed - {err}");
- errors.fetch_add(1, Ordering::SeqCst);
- rename_corrupted_chunk(verify_worker.datastore.clone(), &info.digest);
+ // we must always recheck this here, the parallel worker below alter it!
+ if skip_chunk(&info.digest) {
+ continue; // already verified or marked corrupt
}
- Ok(chunk) => {
- let size = info.size();
- read_bytes += chunk.raw_size();
- decoder_pool.send((chunk, info.digest, size))?;
- decoded_bytes += size;
+
+ match self.datastore.load_chunk(&info.digest) {
+ Err(err) => {
+ self.corrupt_chunks.lock().unwrap().insert(info.digest);
+ error!("can't verify chunk, load failed - {err}");
+ errors.fetch_add(1, Ordering::SeqCst);
+ Self::rename_corrupted_chunk(self.datastore.clone(), &info.digest);
+ }
+ Ok(chunk) => {
+ let size = info.size();
+ read_bytes += chunk.raw_size();
+ decoder_pool.send((chunk, info.digest, size))?;
+ decoded_bytes += size;
+ }
}
}
- }
- decoder_pool.complete()?;
+ decoder_pool.complete()?;
+
+ let elapsed = start_time.elapsed().as_secs_f64();
- let elapsed = start_time.elapsed().as_secs_f64();
+ let read_bytes_mib = (read_bytes as f64) / (1024.0 * 1024.0);
+ let decoded_bytes_mib = (decoded_bytes as f64) / (1024.0 * 1024.0);
- let read_bytes_mib = (read_bytes as f64) / (1024.0 * 1024.0);
- let decoded_bytes_mib = (decoded_bytes as f64) / (1024.0 * 1024.0);
+ let read_speed = read_bytes_mib / elapsed;
+ let decode_speed = decoded_bytes_mib / elapsed;
- let read_speed = read_bytes_mib / elapsed;
- let decode_speed = decoded_bytes_mib / elapsed;
+ let error_count = errors.load(Ordering::SeqCst);
- let error_count = errors.load(Ordering::SeqCst);
+ info!(
+ " verified {read_bytes_mib:.2}/{decoded_bytes_mib:.2} MiB in {elapsed:.2} seconds, speed {read_speed:.2}/{decode_speed:.2} MiB/s ({error_count} errors)"
+ );
- info!(
- " verified {read_bytes_mib:.2}/{decoded_bytes_mib:.2} MiB in {elapsed:.2} seconds, speed {read_speed:.2}/{decode_speed:.2} MiB/s ({error_count} errors)"
- );
+ if errors.load(Ordering::SeqCst) > 0 {
+ bail!("chunks could not be verified");
+ }
- if errors.load(Ordering::SeqCst) > 0 {
- bail!("chunks could not be verified");
+ Ok(())
}
- Ok(())
-}
+ fn verify_fixed_index(&self, backup_dir: &BackupDir, info: &FileInfo) -> Result<(), Error> {
+ let mut path = backup_dir.relative_path();
+ path.push(&info.filename);
-fn verify_fixed_index(
- verify_worker: &VerifyWorker,
- backup_dir: &BackupDir,
- info: &FileInfo,
-) -> Result<(), Error> {
- let mut path = backup_dir.relative_path();
- path.push(&info.filename);
+ let index = self.datastore.open_fixed_reader(&path)?;
- let index = verify_worker.datastore.open_fixed_reader(&path)?;
+ let (csum, size) = index.compute_csum();
+ if size != info.size {
+ bail!("wrong size ({} != {})", info.size, size);
+ }
- let (csum, size) = index.compute_csum();
- if size != info.size {
- bail!("wrong size ({} != {})", info.size, size);
- }
+ if csum != info.csum {
+ bail!("wrong index checksum");
+ }
- if csum != info.csum {
- bail!("wrong index checksum");
+ self.verify_index_chunks(Box::new(index), info.chunk_crypt_mode())
}
- verify_index_chunks(verify_worker, Box::new(index), info.chunk_crypt_mode())
-}
-
-fn verify_dynamic_index(
- verify_worker: &VerifyWorker,
- backup_dir: &BackupDir,
- info: &FileInfo,
-) -> Result<(), Error> {
- let mut path = backup_dir.relative_path();
- path.push(&info.filename);
+ fn verify_dynamic_index(&self, backup_dir: &BackupDir, info: &FileInfo) -> Result<(), Error> {
+ let mut path = backup_dir.relative_path();
+ path.push(&info.filename);
- let index = verify_worker.datastore.open_dynamic_reader(&path)?;
+ let index = self.datastore.open_dynamic_reader(&path)?;
- let (csum, size) = index.compute_csum();
- if size != info.size {
- bail!("wrong size ({} != {})", info.size, size);
- }
-
- if csum != info.csum {
- bail!("wrong index checksum");
- }
+ let (csum, size) = index.compute_csum();
+ if size != info.size {
+ bail!("wrong size ({} != {})", info.size, size);
+ }
- verify_index_chunks(verify_worker, Box::new(index), info.chunk_crypt_mode())
-}
+ if csum != info.csum {
+ bail!("wrong index checksum");
+ }
-/// Verify a single backup snapshot
-///
-/// This checks all archives inside a backup snapshot.
-/// Errors are logged to the worker log.
-///
-/// Returns
-/// - Ok(true) if verify is successful
-/// - Ok(false) if there were verification errors
-/// - Err(_) if task was aborted
-pub fn verify_backup_dir(
- verify_worker: &VerifyWorker,
- backup_dir: &BackupDir,
- upid: UPID,
- filter: Option<&dyn Fn(&BackupManifest) -> bool>,
-) -> Result<bool, Error> {
- if !backup_dir.full_path().exists() {
- info!(
- "SKIPPED: verify {}:{} - snapshot does not exist (anymore).",
- verify_worker.datastore.name(),
- backup_dir.dir(),
- );
- return Ok(true);
+ self.verify_index_chunks(Box::new(index), info.chunk_crypt_mode())
}
- let snap_lock = backup_dir.lock_shared();
-
- match snap_lock {
- Ok(snap_lock) => {
- verify_backup_dir_with_lock(verify_worker, backup_dir, upid, filter, snap_lock)
- }
- Err(err) => {
+ /// Verify a single backup snapshot
+ ///
+ /// This checks all archives inside a backup snapshot.
+ /// Errors are logged to the worker log.
+ ///
+ /// Returns
+ /// - Ok(true) if verify is successful
+ /// - Ok(false) if there were verification errors
+ /// - Err(_) if task was aborted
+ pub fn verify_backup_dir(
+ &self,
+ backup_dir: &BackupDir,
+ upid: UPID,
+ filter: Option<&dyn Fn(&BackupManifest) -> bool>,
+ ) -> Result<bool, Error> {
+ if !backup_dir.full_path().exists() {
info!(
- "SKIPPED: verify {}:{} - could not acquire snapshot lock: {}",
- verify_worker.datastore.name(),
+ "SKIPPED: verify {}:{} - snapshot does not exist (anymore).",
+ self.datastore.name(),
backup_dir.dir(),
- err,
);
- Ok(true)
+ return Ok(true);
}
- }
-}
-/// See verify_backup_dir
-pub fn verify_backup_dir_with_lock(
- verify_worker: &VerifyWorker,
- backup_dir: &BackupDir,
- upid: UPID,
- filter: Option<&dyn Fn(&BackupManifest) -> bool>,
- _snap_lock: BackupLockGuard,
-) -> Result<bool, Error> {
- let datastore_name = verify_worker.datastore.name();
- let backup_dir_name = backup_dir.dir();
-
- let manifest = match backup_dir.load_manifest() {
- Ok((manifest, _)) => manifest,
- Err(err) => {
- info!("verify {datastore_name}:{backup_dir_name} - manifest load error: {err}");
- return Ok(false);
- }
- };
+ let snap_lock = backup_dir.lock_shared();
- if let Some(filter) = filter {
- if !filter(&manifest) {
- info!("SKIPPED: verify {datastore_name}:{backup_dir_name} (recently verified)");
- return Ok(true);
+ match snap_lock {
+ Ok(snap_lock) => self.verify_backup_dir_with_lock(backup_dir, upid, filter, snap_lock),
+ Err(err) => {
+ info!(
+ "SKIPPED: verify {}:{} - could not acquire snapshot lock: {}",
+ self.datastore.name(),
+ backup_dir.dir(),
+ err,
+ );
+ Ok(true)
+ }
}
}
- info!("verify {datastore_name}:{backup_dir_name}");
-
- let mut error_count = 0;
+ /// See verify_backup_dir
+ pub fn verify_backup_dir_with_lock(
+ &self,
+ backup_dir: &BackupDir,
+ upid: UPID,
+ filter: Option<&dyn Fn(&BackupManifest) -> bool>,
+ _snap_lock: BackupLockGuard,
+ ) -> Result<bool, Error> {
+ let datastore_name = self.datastore.name();
+ let backup_dir_name = backup_dir.dir();
+
+ let manifest = match backup_dir.load_manifest() {
+ Ok((manifest, _)) => manifest,
+ Err(err) => {
+ info!("verify {datastore_name}:{backup_dir_name} - manifest load error: {err}");
+ return Ok(false);
+ }
+ };
- let mut verify_result = VerifyState::Ok;
- for info in manifest.files() {
- let result = proxmox_lang::try_block!({
- info!(" check {}", info.filename);
- match ArchiveType::from_path(&info.filename)? {
- ArchiveType::FixedIndex => verify_fixed_index(verify_worker, backup_dir, info),
- ArchiveType::DynamicIndex => verify_dynamic_index(verify_worker, backup_dir, info),
- ArchiveType::Blob => verify_blob(backup_dir, info),
+ if let Some(filter) = filter {
+ if !filter(&manifest) {
+ info!("SKIPPED: verify {datastore_name}:{backup_dir_name} (recently verified)");
+ return Ok(true);
}
- });
+ }
- verify_worker.worker.check_abort()?;
- verify_worker.worker.fail_on_shutdown()?;
+ info!("verify {datastore_name}:{backup_dir_name}");
- if let Err(err) = result {
- info!(
- "verify {datastore_name}:{backup_dir_name}/{file_name} failed: {err}",
- file_name = info.filename,
- );
- error_count += 1;
- verify_result = VerifyState::Failed;
- }
- }
+ let mut error_count = 0;
- let verify_state = SnapshotVerifyState {
- state: verify_result,
- upid,
- };
-
- if let Err(err) = {
- let verify_state = serde_json::to_value(verify_state)?;
- backup_dir.update_manifest(|manifest| {
- manifest.unprotected["verify_state"] = verify_state;
- })
- } {
- info!("verify {datastore_name}:{backup_dir_name} - manifest update error: {err}");
- return Ok(false);
- }
+ let mut verify_result = VerifyState::Ok;
+ for info in manifest.files() {
+ let result = proxmox_lang::try_block!({
+ info!(" check {}", info.filename);
+ match ArchiveType::from_path(&info.filename)? {
+ ArchiveType::FixedIndex => self.verify_fixed_index(backup_dir, info),
+ ArchiveType::DynamicIndex => self.verify_dynamic_index(backup_dir, info),
+ ArchiveType::Blob => Self::verify_blob(backup_dir, info),
+ }
+ });
- Ok(error_count == 0)
-}
+ self.worker.check_abort()?;
+ self.worker.fail_on_shutdown()?;
-/// Verify all backups inside a backup group
-///
-/// Errors are logged to the worker log.
-///
-/// Returns
-/// - Ok((count, failed_dirs)) where failed_dirs had verification errors
-/// - Err(_) if task was aborted
-pub fn verify_backup_group(
- verify_worker: &VerifyWorker,
- group: &BackupGroup,
- progress: &mut StoreProgress,
- upid: &UPID,
- filter: Option<&dyn Fn(&BackupManifest) -> bool>,
-) -> Result<Vec<String>, Error> {
- let mut errors = Vec::new();
- let mut list = match group.list_backups() {
- Ok(list) => list,
- Err(err) => {
- info!(
- "verify {}, group {} - unable to list backups: {}",
- print_store_and_ns(verify_worker.datastore.name(), group.backup_ns()),
- group.group(),
- err,
- );
- return Ok(errors);
- }
- };
-
- let snapshot_count = list.len();
- info!(
- "verify group {}:{} ({} snapshots)",
- verify_worker.datastore.name(),
- group.group(),
- snapshot_count
- );
-
- progress.group_snapshots = snapshot_count as u64;
-
- BackupInfo::sort_list(&mut list, false); // newest first
- for (pos, info) in list.into_iter().enumerate() {
- if !verify_backup_dir(verify_worker, &info.backup_dir, upid.clone(), filter)? {
- errors.push(print_ns_and_snapshot(
- info.backup_dir.backup_ns(),
- info.backup_dir.as_ref(),
- ));
+ if let Err(err) = result {
+ info!(
+ "verify {datastore_name}:{backup_dir_name}/{file_name} failed: {err}",
+ file_name = info.filename,
+ );
+ error_count += 1;
+ verify_result = VerifyState::Failed;
+ }
}
- progress.done_snapshots = pos as u64 + 1;
- info!("percentage done: {progress}");
- }
- Ok(errors)
-}
-/// Verify all (owned) backups inside a datastore
-///
-/// Errors are logged to the worker log.
-///
-/// Returns
-/// - Ok(failed_dirs) where failed_dirs had verification errors
-/// - Err(_) if task was aborted
-pub fn verify_all_backups(
- verify_worker: &VerifyWorker,
- upid: &UPID,
- ns: BackupNamespace,
- max_depth: Option<usize>,
- owner: Option<&Authid>,
- filter: Option<&dyn Fn(&BackupManifest) -> bool>,
-) -> Result<Vec<String>, Error> {
- let mut errors = Vec::new();
-
- info!("verify datastore {}", verify_worker.datastore.name());
-
- let owner_filtered = if let Some(owner) = &owner {
- info!("limiting to backups owned by {owner}");
- true
- } else {
- false
- };
-
- // FIXME: This should probably simply enable recursion (or the call have a recursion parameter)
- let store = &verify_worker.datastore;
- let max_depth = max_depth.unwrap_or(pbs_api_types::MAX_NAMESPACE_DEPTH);
-
- let mut list = match ListAccessibleBackupGroups::new_with_privs(
- store,
- ns.clone(),
- max_depth,
- Some(PRIV_DATASTORE_VERIFY),
- Some(PRIV_DATASTORE_BACKUP),
- owner,
- ) {
- Ok(list) => list
- .filter_map(|group| match group {
- Ok(group) => Some(group),
- Err(err) if owner_filtered => {
- // intentionally not in task log, the user might not see this group!
- println!("error on iterating groups in ns '{ns}' - {err}");
- None
- }
- Err(err) => {
- // we don't filter by owner, but we want to log the error
- info!("error on iterating groups in ns '{ns}' - {err}");
- errors.push(err.to_string());
- None
- }
- })
- .filter(|group| {
- !(group.backup_type() == BackupType::Host && group.backup_id() == "benchmark")
+ let verify_state = SnapshotVerifyState {
+ state: verify_result,
+ upid,
+ };
+
+ if let Err(err) = {
+ let verify_state = serde_json::to_value(verify_state)?;
+ backup_dir.update_manifest(|manifest| {
+ manifest.unprotected["verify_state"] = verify_state;
})
- .collect::<Vec<BackupGroup>>(),
- Err(err) => {
- info!("unable to list backups: {err}");
- return Ok(errors);
+ } {
+ info!("verify {datastore_name}:{backup_dir_name} - manifest update error: {err}");
+ return Ok(false);
}
- };
- list.sort_unstable_by(|a, b| a.group().cmp(b.group()));
+ Ok(error_count == 0)
+ }
- let group_count = list.len();
- info!("found {group_count} groups");
+ /// Verify all backups inside a backup group
+ ///
+ /// Errors are logged to the worker log.
+ ///
+ /// Returns
+ /// - Ok((count, failed_dirs)) where failed_dirs had verification errors
+ /// - Err(_) if task was aborted
+ pub fn verify_backup_group(
+ &self,
+ group: &BackupGroup,
+ progress: &mut StoreProgress,
+ upid: &UPID,
+ filter: Option<&dyn Fn(&BackupManifest) -> bool>,
+ ) -> Result<Vec<String>, Error> {
+ let mut errors = Vec::new();
+ let mut list = match group.list_backups() {
+ Ok(list) => list,
+ Err(err) => {
+ info!(
+ "verify {}, group {} - unable to list backups: {}",
+ print_store_and_ns(self.datastore.name(), group.backup_ns()),
+ group.group(),
+ err,
+ );
+ return Ok(errors);
+ }
+ };
- let mut progress = StoreProgress::new(group_count as u64);
+ let snapshot_count = list.len();
+ info!(
+ "verify group {}:{} ({} snapshots)",
+ self.datastore.name(),
+ group.group(),
+ snapshot_count
+ );
- for (pos, group) in list.into_iter().enumerate() {
- progress.done_groups = pos as u64;
- progress.done_snapshots = 0;
- progress.group_snapshots = 0;
+ progress.group_snapshots = snapshot_count as u64;
- let mut group_errors =
- verify_backup_group(verify_worker, &group, &mut progress, upid, filter)?;
- errors.append(&mut group_errors);
+ BackupInfo::sort_list(&mut list, false); // newest first
+ for (pos, info) in list.into_iter().enumerate() {
+ if !self.verify_backup_dir(&info.backup_dir, upid.clone(), filter)? {
+ errors.push(print_ns_and_snapshot(
+ info.backup_dir.backup_ns(),
+ info.backup_dir.as_ref(),
+ ));
+ }
+ progress.done_snapshots = pos as u64 + 1;
+ info!("percentage done: {progress}");
+ }
+ Ok(errors)
}
- Ok(errors)
-}
+ /// Verify all (owned) backups inside a datastore
+ ///
+ /// Errors are logged to the worker log.
+ ///
+ /// Returns
+ /// - Ok(failed_dirs) where failed_dirs had verification errors
+ /// - Err(_) if task was aborted
+ pub fn verify_all_backups(
+ &self,
+ upid: &UPID,
+ ns: BackupNamespace,
+ max_depth: Option<usize>,
+ owner: Option<&Authid>,
+ filter: Option<&dyn Fn(&BackupManifest) -> bool>,
+ ) -> Result<Vec<String>, Error> {
+ let mut errors = Vec::new();
+
+ info!("verify datastore {}", self.datastore.name());
+
+ let owner_filtered = if let Some(owner) = &owner {
+ info!("limiting to backups owned by {owner}");
+ true
+ } else {
+ false
+ };
+
+ // FIXME: This should probably simply enable recursion (or the call have a recursion parameter)
+ let store = &self.datastore;
+ let max_depth = max_depth.unwrap_or(pbs_api_types::MAX_NAMESPACE_DEPTH);
+
+ let mut list = match ListAccessibleBackupGroups::new_with_privs(
+ store,
+ ns.clone(),
+ max_depth,
+ Some(PRIV_DATASTORE_VERIFY),
+ Some(PRIV_DATASTORE_BACKUP),
+ owner,
+ ) {
+ Ok(list) => list
+ .filter_map(|group| match group {
+ Ok(group) => Some(group),
+ Err(err) if owner_filtered => {
+ // intentionally not in task log, the user might not see this group!
+ println!("error on iterating groups in ns '{ns}' - {err}");
+ None
+ }
+ Err(err) => {
+ // we don't filter by owner, but we want to log the error
+ info!("error on iterating groups in ns '{ns}' - {err}");
+ errors.push(err.to_string());
+ None
+ }
+ })
+ .filter(|group| {
+ !(group.backup_type() == BackupType::Host && group.backup_id() == "benchmark")
+ })
+ .collect::<Vec<BackupGroup>>(),
+ Err(err) => {
+ info!("unable to list backups: {err}");
+ return Ok(errors);
+ }
+ };
+
+ list.sort_unstable_by(|a, b| a.group().cmp(b.group()));
+
+ let group_count = list.len();
+ info!("found {group_count} groups");
-/// Filter out any snapshot from being (re-)verified where this fn returns false.
-pub fn verify_filter(
- ignore_verified_snapshots: bool,
- outdated_after: Option<i64>,
- manifest: &BackupManifest,
-) -> bool {
- if !ignore_verified_snapshots {
- return true;
+ let mut progress = StoreProgress::new(group_count as u64);
+
+ for (pos, group) in list.into_iter().enumerate() {
+ progress.done_groups = pos as u64;
+ progress.done_snapshots = 0;
+ progress.group_snapshots = 0;
+
+ let mut group_errors = self.verify_backup_group(&group, &mut progress, upid, filter)?;
+ errors.append(&mut group_errors);
+ }
+
+ Ok(errors)
}
- match manifest.verify_state() {
- Err(err) => {
- warn!("error reading manifest: {err:#}");
- true
+ /// Filter out any snapshot from being (re-)verified where this fn returns false.
+ pub fn verify_filter(
+ ignore_verified_snapshots: bool,
+ outdated_after: Option<i64>,
+ manifest: &BackupManifest,
+ ) -> bool {
+ if !ignore_verified_snapshots {
+ return true;
}
- Ok(None) => true, // no last verification, always include
- Ok(Some(last_verify)) => {
- match outdated_after {
- None => false, // never re-verify if ignored and no max age
- Some(max_age) => {
- let now = proxmox_time::epoch_i64();
- let days_since_last_verify = (now - last_verify.upid.starttime) / 86400;
-
- days_since_last_verify > max_age
+
+ match manifest.verify_state() {
+ Err(err) => {
+ warn!("error reading manifest: {err:#}");
+ true
+ }
+ Ok(None) => true, // no last verification, always include
+ Ok(Some(last_verify)) => {
+ match outdated_after {
+ None => false, // never re-verify if ignored and no max age
+ Some(max_age) => {
+ let now = proxmox_time::epoch_i64();
+ let days_since_last_verify = (now - last_verify.upid.starttime) / 86400;
+
+ days_since_last_verify > max_age
+ }
}
}
}
diff --git a/src/server/verify_job.rs b/src/server/verify_job.rs
index a15a257da..95a7b2a9b 100644
--- a/src/server/verify_job.rs
+++ b/src/server/verify_job.rs
@@ -5,10 +5,7 @@ use pbs_api_types::{Authid, Operation, VerificationJobConfig};
use pbs_datastore::DataStore;
use proxmox_rest_server::WorkerTask;
-use crate::{
- backup::{verify_all_backups, verify_filter},
- server::jobstate::Job,
-};
+use crate::{backup::VerifyWorker, server::jobstate::Job};
/// Runs a verification job.
pub fn do_verification_job(
@@ -44,15 +41,14 @@ pub fn do_verification_job(
None => Default::default(),
};
- let verify_worker = crate::backup::VerifyWorker::new(worker.clone(), datastore);
- let result = verify_all_backups(
- &verify_worker,
+ let verify_worker = VerifyWorker::new(worker.clone(), datastore);
+ let result = verify_worker.verify_all_backups(
worker.upid(),
ns,
verification_job.max_depth,
None,
Some(&move |manifest| {
- verify_filter(ignore_verified_snapshots, outdated_after, manifest)
+ VerifyWorker::verify_filter(ignore_verified_snapshots, outdated_after, manifest)
}),
);
let job_result = match result {
--
2.39.5
More information about the pbs-devel
mailing list