crimson/osd/osd: handle_scrub_message on any core

After enabling multicore cluster messenger, there is no
guarantee which core will handle the request.
See: OSDSingletonState::send_to_osd()

Signed-off-by: Matan Breizman <mbreizma@redhat.com>
(cherry picked from commit cbad3e4517e91476b19a8766f98240c81df40a36)

diff --git a/src/crimson/osd/osd.cc b/src/crimson/osd/osd.cc
index 5c02b2895c2d9f3240cce2dbb24285a80f7f342b..939fbc59beb8cd6f820831550944bfba9dfcd2f2 100644 (file)
--- a/src/crimson/osd/osd.cc
+++ b/src/crimson/osd/osd.cc
@@ -1268,7 +1268,6 @@ seastar::future<> OSD::handle_scrub_message(
   crimson::net::ConnectionRef conn,
   Ref<MOSDFastDispatchOp> m)
 {
-  ceph_assert(seastar::this_shard_id() == PRIMARY_CORE);
   return pg_shard_manager.start_pg_operation<
     crimson::osd::ScrubMessage
     >(m, conn, m->get_min_epoch(), m->get_spg()).second;