From: Patrick Donnelly <pdonnell@ibm.com>
Date: Wed, 29 Jan 2025 20:26:45 +0000 (-0500)
Subject: qa: avoid unnecessary mds restart
X-Git-Tag: v20.0.0~226^2~10
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=cb3229bf2fb74b097efe721f08ce43f56a03e2ee;p=ceph.git

qa: avoid unnecessary mds restart

The MDS are already sitting in standby because the fs is damaged.

Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
---

diff --git a/qa/tasks/cephfs/test_data_scan.py b/qa/tasks/cephfs/test_data_scan.py
index 30be5c1a47dc..6c75d2144579 100644
--- a/qa/tasks/cephfs/test_data_scan.py
+++ b/qa/tasks/cephfs/test_data_scan.py
@@ -416,8 +416,8 @@ class TestDataScan(CephFSTestCase):
 
         # After recovery, we need the MDS to not be strict about stats (in production these options
         # are off by default, but in QA we need to explicitly disable them)
-        self.fs.set_ceph_conf('mds', 'mds verify scatter', False)
-        self.fs.set_ceph_conf('mds', 'mds debug scatterstat', False)
+        self.config_set('mds', 'mds verify scatter', False)
+        self.config_set('mds', 'mds debug scatterstat', False)
 
         # Apply any data damage the workload wants
         workload.damage()
@@ -461,7 +461,6 @@ class TestDataScan(CephFSTestCase):
         self.run_ceph_cmd('mds', 'repaired', '0')
 
         # Start the MDS
-        self.fs.mds_restart()
         self.fs.wait_for_daemons()
         log.info(str(self.mds_cluster.status()))