From: Sage Weil <sage@redhat.com>
Date: Thu, 24 Aug 2017 21:56:13 +0000 (-0400)
Subject: osd: subscribe to new osdmap while waiting_for_healthy
X-Git-Tag: v12.2.1~88^2
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=refs%2Fpull%2F17498%2Fhead;p=ceph.git

osd: subscribe to new osdmap while waiting_for_healthy

If we are sitting around waiting until we are able to ping our "up" peers,
we need to be sure that our notion of "up" is still correct and we're not
just stuck on an old, stale OSDMap.

Fixes: http://tracker.ceph.com/issues/21121
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit fbafa659dae94faba435ca449ee5e77b51108b4b)
---

diff --git a/src/osd/OSD.cc b/src/osd/OSD.cc
index 28608f1250a9..d8ce26b12947 100644
--- a/src/osd/OSD.cc
+++ b/src/osd/OSD.cc
@@ -5785,6 +5785,9 @@ void OSD::start_waiting_for_healthy()
   dout(1) << "start_waiting_for_healthy" << dendl;
   set_state(STATE_WAITING_FOR_HEALTHY);
   last_heartbeat_resample = utime_t();
+
+  // subscribe to osdmap updates, in case our peers really are known to be dead
+  osdmap_subscribe(osdmap->get_epoch() + 1, false);
 }
 
 bool OSD::_is_healthy()