[v3,13/16] qcow2: Repair snapshot table with too many entries
diff mbox series

Message ID 20191011152814.14791-14-mreitz@redhat.com
State New
Headers show
Series
  • qcow2: Let check -r all repair some snapshot bits
Related show

Commit Message

Max Reitz Oct. 11, 2019, 3:28 p.m. UTC
The user cannot choose which snapshots are removed.  This is fine
because we have chosen the maximum snapshot table size to be so large
(65536 entries) that it cannot be reasonably reached.  If the snapshot
table exceeds this size, the image has probably been corrupted in some
way; in this case, it is most important to just make the image usable
such that the user can copy off at least the active layer.
(Also note that the snapshots will be removed only with "-r all", so a
plain "check" or "check -r leaks" will not delete any data.)

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
---
 block/qcow2-snapshot.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

Comments

Eric Blake Oct. 11, 2019, 4:31 p.m. UTC | #1
On 10/11/19 10:28 AM, Max Reitz wrote:
> The user cannot choose which snapshots are removed.  This is fine
> because we have chosen the maximum snapshot table size to be so large
> (65536 entries) that it cannot be reasonably reached.  If the snapshot
> table exceeds this size, the image has probably been corrupted in some
> way; in this case, it is most important to just make the image usable
> such that the user can copy off at least the active layer.
> (Also note that the snapshots will be removed only with "-r all", so a
> plain "check" or "check -r leaks" will not delete any data.)
> 
> Signed-off-by: Max Reitz <mreitz@redhat.com>
> Reviewed-by: Eric Blake <eblake@redhat.com>

The updated commit messages are good.

Patch
diff mbox series

diff --git a/block/qcow2-snapshot.c b/block/qcow2-snapshot.c
index 366d9f574c..dac8a778e4 100644
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -444,6 +444,14 @@  int coroutine_fn qcow2_check_read_snapshot_table(BlockDriverState *bs,
     s->snapshots_offset = be64_to_cpu(snapshot_table_pointer.snapshots_offset);
     s->nb_snapshots = be32_to_cpu(snapshot_table_pointer.nb_snapshots);
 
+    if (s->nb_snapshots > QCOW_MAX_SNAPSHOTS && (fix & BDRV_FIX_ERRORS)) {
+        fprintf(stderr, "Discarding %u overhanging snapshots\n",
+                s->nb_snapshots - QCOW_MAX_SNAPSHOTS);
+
+        nb_clusters_reduced += s->nb_snapshots - QCOW_MAX_SNAPSHOTS;
+        s->nb_snapshots = QCOW_MAX_SNAPSHOTS;
+    }
+
     ret = qcow2_validate_table(bs, s->snapshots_offset, s->nb_snapshots,
                                sizeof(QCowSnapshotHeader),
                                sizeof(QCowSnapshotHeader) * QCOW_MAX_SNAPSHOTS,
@@ -452,6 +460,12 @@  int coroutine_fn qcow2_check_read_snapshot_table(BlockDriverState *bs,
         result->check_errors++;
         error_reportf_err(local_err, "ERROR ");
 
+        if (s->nb_snapshots > QCOW_MAX_SNAPSHOTS) {
+            fprintf(stderr, "You can force-remove all %u overhanging snapshots "
+                    "with qemu-img check -r all\n",
+                    s->nb_snapshots - QCOW_MAX_SNAPSHOTS);
+        }
+
         /* We did not read the snapshot table, so invalidate this information */
         s->snapshots_offset = 0;
         s->nb_snapshots = 0;