Merge pull request #1239 from ThomasWaldmann/forced-archive-delete

Forced archive delete

borg/archive.py

@@ -24,6 +24,8 @@ from .helpers import Error, uid2user, user2uid, gid2group, group2gid, \
 from .platform import acl_get, acl_set
 from .chunker import Chunker
 from .hashindex import ChunkIndex
+from .repository import Repository
+
 import msgpack
 
 ITEMS_BUFFER = 1024 * 1024
@@ -494,23 +496,59 @@ Number of files: {0.stats.nfiles}'''.format(
         self.cache.chunk_decref(self.id, self.stats)
         del self.manifest.archives[self.name]
 
-    def delete(self, stats, progress=False):
-        unpacker = msgpack.Unpacker(use_list=False)
-        items_ids = self.metadata[b'items']
-        pi = ProgressIndicatorPercent(total=len(items_ids), msg="Decrementing references %3.0f%%", same_line=True)
-        for (i, (items_id, data)) in enumerate(zip(items_ids, self.repository.get_many(items_ids))):
+    def delete(self, stats, progress=False, forced=False):
+        class ChunksIndexError(Error):
+            """Chunk ID {} missing from chunks index, corrupted chunks index - aborting transaction."""
+
+        def chunk_decref(id, stats):
+            nonlocal error
+            try:
+                self.cache.chunk_decref(id, stats)
+            except KeyError:
+                cid = hexlify(id).decode('ascii')
+                raise ChunksIndexError(cid)
+            except Repository.ObjectNotFound as e:
+                # object not in repo - strange, but we wanted to delete it anyway.
+                if not forced:
+                    raise
+                error = True
+
+        error = False
+        try:
+            unpacker = msgpack.Unpacker(use_list=False)
+            items_ids = self.metadata[b'items']
+            pi = ProgressIndicatorPercent(total=len(items_ids), msg="Decrementing references %3.0f%%", same_line=True)
+            for (i, (items_id, data)) in enumerate(zip(items_ids, self.repository.get_many(items_ids))):
+                if progress:
+                    pi.show(i)
+                unpacker.feed(self.key.decrypt(items_id, data))
+                chunk_decref(items_id, stats)
+                try:
+                    for item in unpacker:
+                        if b'chunks' in item:
+                            for chunk_id, size, csize in item[b'chunks']:
+                                chunk_decref(chunk_id, stats)
+                except (TypeError, ValueError):
+                    # if items metadata spans multiple chunks and one chunk got dropped somehow,
+                    # it could be that unpacker yields bad types
+                    if not forced:
+                        raise
+                    error = True
             if progress:
-                pi.show(i)
-            unpacker.feed(self.key.decrypt(items_id, data))
-            self.cache.chunk_decref(items_id, stats)
-            for item in unpacker:
-                if b'chunks' in item:
-                    for chunk_id, size, csize in item[b'chunks']:
-                        self.cache.chunk_decref(chunk_id, stats)
-        if progress:
-            pi.finish()
-        self.cache.chunk_decref(self.id, stats)
+                pi.finish()
+        except (msgpack.UnpackException, Repository.ObjectNotFound):
+            # items metadata corrupted
+            if not forced:
+                raise
+            error = True
+        # in forced delete mode, we try hard to delete at least the manifest entry,
+        # if possible also the archive superblock, even if processing the items raises
+        # some harmless exception.
+        chunk_decref(self.id, stats)
         del self.manifest.archives[self.name]
+        if error:
+            logger.warning('forced deletion succeeded, but the deleted archive was corrupted.')
+            logger.warning('borg check --repair is required to free all space.')
 
     def stat_attrs(self, st, path):
         item = {

borg/archiver.py

@@ -426,7 +426,7 @@ class Archiver:
             with Cache(repository, key, manifest, lock_wait=self.lock_wait) as cache:
                 archive = Archive(repository, key, manifest, args.location.archive, cache=cache)
                 stats = Statistics()
-                archive.delete(stats, progress=args.progress)
+                archive.delete(stats, progress=args.progress, forced=args.forced)
                 manifest.write()
                 repository.commit(save_space=args.save_space)
                 cache.commit()
@@ -635,7 +635,7 @@ class Archiver:
                 else:
                     if args.output_list:
                         logger.info('Pruning archive: %s' % format_archive(archive))
-                    Archive(repository, key, manifest, archive.name, cache).delete(stats)
+                    Archive(repository, key, manifest, archive.name, cache).delete(stats, forced=args.forced)
             if to_delete and not args.dry_run:
                 manifest.write()
                 repository.commit(save_space=args.save_space)
@@ -1230,6 +1230,9 @@ class Archiver:
         subparser.add_argument('-c', '--cache-only', dest='cache_only',
                                action='store_true', default=False,
                                help='delete only the local cache for the given repository')
+        subparser.add_argument('--force', dest='forced',
+                               action='store_true', default=False,
+                               help='force deletion of corrupted archives')
         subparser.add_argument('--save-space', dest='save_space', action='store_true',
                                default=False,
                                help='work slower, but using less space')
@@ -1342,6 +1345,9 @@ class Archiver:
         subparser.add_argument('-n', '--dry-run', dest='dry_run',
                                default=False, action='store_true',
                                help='do not change repository')
+        subparser.add_argument('--force', dest='forced',
+                               action='store_true', default=False,
+                               help='force pruning of corrupted archives')
         subparser.add_argument('-s', '--stats', dest='stats',
                                action='store_true', default=False,
                                help='print statistics for the deleted archive')