rgw: limit entries in remove_olh_pending_entries()

If there are too many entries to send in a single osd op, the osd rejects
the request with EINVAL. This error happens in follow_olh(), which means
that requests against the object logical head (requests with no version
id) can't be resolved to the current object version. In multisite, this
also causes data sync to get stuck in retries

Fixes: http://tracker.ceph.com/issues/39118
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 3805ea635a5f28656072b2cf5cdc9b5bf53b5e97)

Conflicts:
	src/rgw/rgw_rados.cc
- mimic ref.ioctx.operate() call takes ref.oid instead of ref.obj.oid

diff --git a/src/rgw/rgw_rados.cc b/src/rgw/rgw_rados.cc
index 88e870054031f0fefe59080e6500f78982a65790..c6bd3e9f35ebb4ccde1d3131afe1adca443f421b 100644 (file)
--- a/src/rgw/rgw_rados.cc
+++ b/src/rgw/rgw_rados.cc
@@ -11870,31 +11870,34 @@ void RGWRados::check_pending_olh_entries(map<string, bufferlist>& pending_entrie
 
 int RGWRados::remove_olh_pending_entries(const RGWBucketInfo& bucket_info, RGWObjState& state, const rgw_obj& olh_obj, map<string, bufferlist>& pending_attrs)
 {
-  ObjectWriteOperation op;
-
-  bucket_index_guard_olh_op(state, op);
-
-  for (map<string, bufferlist>::iterator iter = pending_attrs.begin(); iter != pending_attrs.end(); ++iter) {
-    op.rmxattr(iter->first.c_str());
-  }
-
   rgw_rados_ref ref;
   int r = get_obj_head_ref(bucket_info, olh_obj, &ref);
   if (r < 0) {
     return r;
   }
 
-  /* update olh object */
-  r = ref.ioctx.operate(ref.oid, &op);
-  if (r == -ENOENT || r == -ECANCELED) {
-    /* raced with some other change, shouldn't sweat about it */
-    r = 0;
-  }
-  if (r < 0) {
-    ldout(cct, 0) << "ERROR: could not apply olh update, r=" << r << dendl;
-    return r;
-  }
+  // trim no more than 1000 entries per osd op
+  constexpr int max_entries = 1000;
+
+  auto i = pending_attrs.begin();
+  while (i != pending_attrs.end()) {
+    ObjectWriteOperation op;
+    bucket_index_guard_olh_op(state, op);
 
+    for (int n = 0; n < max_entries && i != pending_attrs.end(); ++n, ++i) {
+      op.rmxattr(i->first.c_str());
+    }
+
+    r = ref.ioctx.operate(ref.oid, &op);
+    if (r == -ENOENT || r == -ECANCELED) {
+      /* raced with some other change, shouldn't sweat about it */
+      return 0;
+    }
+    if (r < 0) {
+      ldout(cct, 0) << "ERROR: could not apply olh update, r=" << r << dendl;
+      return r;
+    }
+  }
   return 0;
 }