Fix possible race condition in swift_migration check

When the replicator is still moving data off the standalone node, the
find command might fail with a race condition because empty directories
are removed while iterating over the same subdirectories.

This fix adds the "-ignore_readdir_race" option to find to ignore this
kind of errors. The test should simply succeed as long as there are no
.data or .db files left; empty directories will be cleaned up later and
are ok.

JIRA: OSPCIX-1318

Signed-off-by: Christian Schwede <cschwede@redhat.com>

Author: cschwede

tests/roles/swift_migration/tasks/main.yaml

@@ -57,7 +57,7 @@
   ansible.builtin.shell: |
     {{ shell_header }}
     CONTROLLER1_SSH="{{ controller1_ssh }}"
-    $CONTROLLER1_SSH "timeout 900s bash -c 'while \$(find /srv/node/ -type f -name \"*.db\" -o -name \"*.data\" | grep -q \".\"); do sleep 5; done'"
+    $CONTROLLER1_SSH "timeout 900s bash -c 'while \$(find /srv/node/ -ignore_readdir_race -type f -name \"*.db\" -o -name \"*.data\" | grep -q \".\"); do sleep 5; done'"
 
 - name: remove standalone node from rings
   ansible.builtin.shell: |