osd: Fix incorrect rollback logic for partial write OI

Before this fix, when rolling back an OI, the system used the OI
from the primary to rollback to. This is wrong if the previous
write was a partial write. This may have a few consequences
during recovery (although its not clear any are serious) and in
EC direct reads, where a false-positive version mismatch will be
detected.

The test provided recreates the issue.

The fix provided modifies the rollback as it is being written.

Fixes: https://tracker.ceph.com/issues/76213
Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>
# Conflicts:
#	src/test/osd/TestECFailoverWithPeering.cc

diff --git a/src/osd/PGBackend.cc b/src/osd/PGBackend.cc
index e17f4f3909d442d25688f7a6a60b12bd59da1eb6..bb40c28e66c9179e90290d08e475a43d41f7009a 100644 (file)
--- a/src/osd/PGBackend.cc
+++ b/src/osd/PGBackend.cc
@@ -672,11 +672,30 @@ void PGBackend::rollback_setattrs(
     }
   }
   if (only_oi) {
-    t->setattr(
-      coll,
-      ghobject_t(hoid, ghobject_t::NO_GEN, get_parent()->whoami_shard().shard),
-      OI_ATTR,
-      to_set[OI_ATTR]);
+    object_info_t oi;
+    auto p = to_set[OI_ATTR].cbegin();
+    decode(oi, p);
+
+    shard_id_t my_shard = get_parent()->whoami_shard().shard;
+    if (oi.shard_versions.contains(my_shard) && oi.shard_versions.at(my_shard) != oi.version) {
+      oi.version = oi.shard_versions.at(my_shard);
+      oi.shard_versions.clear();
+      
+      bufferlist bl;
+      encode(oi, bl, get_osdmap()->get_features(CEPH_ENTITY_TYPE_OSD, nullptr));
+      
+      t->setattr(
+        coll,
+        ghobject_t(hoid, ghobject_t::NO_GEN, my_shard),
+        OI_ATTR,
+        bl);
+    } else {
+      t->setattr(
+        coll,
+        ghobject_t(hoid, ghobject_t::NO_GEN, my_shard),
+        OI_ATTR,
+        to_set[OI_ATTR]);
+    }
   } else {
     t->setattrs(
       coll,

diff --git a/src/test/osd/TestECFailoverWithPeering.cc b/src/test/osd/TestECFailoverWithPeering.cc
index 2ac088d3975ad15133ca93cc29701a09ac881dab..e06ec7b73ff08241a312ac6d5a9514b8f3d03105 100644 (file)
--- a/src/test/osd/TestECFailoverWithPeering.cc
+++ b/src/test/osd/TestECFailoverWithPeering.cc
@@ -453,7 +453,7 @@ TEST_P(TestECFailoverWithPeering, ECSequentialOSDFailoverTest) {
  *
  * Recreate https://tracker.ceph.com/issues/76213
  */
-TEST_P(TestECFailoverWithPeering, DISABLED_RollbackVersionMismatch) {
+TEST_P(TestECFailoverWithPeering, RollbackVersionMismatch) {
   if (k < 3) {
     GTEST_SKIP() << "SnapshotTrimRollbackVersionMismatch requires at least 3 data shards";
   }