From: Sage Weil <sage@redhat.com>
Date: Tue, 28 Jan 2020 19:33:49 +0000 (-0600)
Subject: osd: dispatch_context and queue split finish on early bail-out
X-Git-Tag: v15.1.0~11^2
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=refs%2Fpull%2F32942%2Fhead;p=ceph.git

osd: dispatch_context and queue split finish on early bail-out

If we bail out of advance_pg early because there is an upcoming merge, we
still need to dispatch_context() on rctx before we drop the PG lock.  And
the rctx that we submit needs to include the on_applied finisher comit
to call _finish_splits.

This is noticeable (at least) when there is a split and merge that are
both known.  When we process the split, the new child is added to new_pgs.
When we get to the merge epoch, we stop early and take the bail-out
path.

Fix by adding a dispatch_context call for this path.  And further make sure
that both dispatch_context callers in this function queue up the
new_pgs event.

Fixes: https://tracker.ceph.com/issues/43825
Signed-off-by: Sage Weil <sage@redhat.com>
---

diff --git a/src/osd/OSD.cc b/src/osd/OSD.cc
index c7eab5587896..6d7119008742 100644
--- a/src/osd/OSD.cc
+++ b/src/osd/OSD.cc
@@ -8465,6 +8465,11 @@ bool OSD::advance_pg(
 		  << " is merge source, target is " << parent
 		   << dendl;
 	  pg->write_if_dirty(rctx);
+	  if (!new_pgs.empty()) {
+	    rctx.transaction.register_on_applied(new C_FinishSplits(this,
+								    new_pgs));
+	    new_pgs.clear();
+	  }
 	  dispatch_context(rctx, pg, pg->get_osdmap(), &handle);
 	  pg->ch->flush();
 	  // release backoffs explicitly, since the on_shutdown path
@@ -8537,6 +8542,12 @@ bool OSD::advance_pg(
 	  } else {
 	    dout(20) << __func__ << " not ready to merge yet" << dendl;
 	    pg->write_if_dirty(rctx);
+	    if (!new_pgs.empty()) {
+	      rctx.transaction.register_on_applied(new C_FinishSplits(this,
+								      new_pgs));
+	      new_pgs.clear();
+	    }
+	    dispatch_context(rctx, pg, pg->get_osdmap(), &handle);
 	    pg->unlock();
 	    // kick source(s) to get them ready
 	    for (auto& i : children) {