qa/tasks/osd_backfill.py: wait for osd.[12] to start

...before sending a tell command.  Otherwise osd.2 might
start without 1, the io unblocks, and the tell fails
because osd.1 is still down.

Fixes: http://tracker.ceph.com/issues/18303
Signed-off-by: Sage Weil <sage@redhat.com>

diff --git a/qa/tasks/osd_backfill.py b/qa/tasks/osd_backfill.py
index f0bba7963eed98763ca8e1891a3083661ee8ee74..dbed30fd412283f4f45fc3d5a5400b1f9c6269e7 100644 (file)
--- a/qa/tasks/osd_backfill.py
+++ b/qa/tasks/osd_backfill.py
@@ -91,6 +91,10 @@ def task(ctx, config):
     err = p.wait()
     log.info('err is %d' % err)
 
+    # wait for osd.1 and osd.2 to be up
+    manager.wait_till_osd_is_up(1)
+    manager.wait_till_osd_is_up(2)
+
     # cluster must recover
     manager.raw_cluster_cmd('tell', 'osd.1', 'flush_pg_stats')
     manager.raw_cluster_cmd('tell', 'osd.2', 'flush_pg_stats')