qa/suites: drop --show-reachable=yes from fs:valgrind tests

Drop --show-reachable=yes from fs:valgrind tests as it shows CommandFailedError,
which fails starting cephfs-mirror daemon and prevents the valgrind checks and reports.
rbd:valgrind tests doesn't use --show-reachable=yes too.
So dropping --show-reachable=yes to make it analogous to rbd:valgrind.

Fixes: https://tracker.ceph.com/issues/64752
Fixes: https://tracker.ceph.com/issues/62221
Signed-off-by: Jos Collin <jcollin@redhat.com>
(cherry picked from commit f17f2d6fee1dc59958678b697a0e0319c34691a7)

diff --git a/qa/suites/fs/valgrind/mirror/cephfs-mirror/one-per-cluster.yaml b/qa/suites/fs/valgrind/mirror/cephfs-mirror/one-per-cluster.yaml
index 4112a0af0b10f8fdcc540e9b5eddf89e3ce255f3..41518f5d1fdec7f35140f7bb9d2fe633a1f06766 100644 (file)
--- a/qa/suites/fs/valgrind/mirror/cephfs-mirror/one-per-cluster.yaml
+++ b/qa/suites/fs/valgrind/mirror/cephfs-mirror/one-per-cluster.yaml
@@ -4,4 +4,4 @@ meta:
 tasks:
 - cephfs-mirror:
     client: client.mirror
-    valgrind: [--tool=memcheck, --leak-check=full, --show-reachable=yes]
+    valgrind: [--tool=memcheck, --leak-check=full]