From: Patrick Donnelly <pdonnell@redhat.com>
Date: Mon, 1 Jul 2019 18:21:13 +0000 (-0700)
Subject: qa: use hard_reset to reboot kclient
X-Git-Tag: v15.1.0~2037^2~1
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=b75544be86ba3300d386d1039fa3059afff085e9;p=ceph.git

qa: use hard_reset to reboot kclient

power_off may allow the mounts to gracefully unmount. We don't want this if the
kclient is stuck or we desire the client to "disappear" and come back.

Fixes: http://tracker.ceph.com/issues/37681
Depends-on: https://github.com/ceph/teuthology/pull/1296
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
---

diff --git a/qa/tasks/cephfs/kernel_mount.py b/qa/tasks/cephfs/kernel_mount.py
index 6b128f57247d..950a1137d1b7 100644
--- a/qa/tasks/cephfs/kernel_mount.py
+++ b/qa/tasks/cephfs/kernel_mount.py
@@ -1,6 +1,7 @@
 from StringIO import StringIO
 import json
 import logging
+import time
 from textwrap import dedent
 from teuthology.orchestra.run import CommandFailedError
 from teuthology import misc
@@ -176,21 +177,31 @@ class KernelMount(CephFSMount):
                                                 self.ipmi_user,
                                                 self.ipmi_password,
                                                 self.ipmi_domain)
-        con.power_off()
+        con.hard_reset(wait_for_login=False)
 
         self.mounted = False
 
     def kill_cleanup(self):
         assert not self.mounted
 
-        con = orchestra_remote.getRemoteConsole(self.client_remote.hostname,
-                                                self.ipmi_user,
-                                                self.ipmi_password,
-                                                self.ipmi_domain)
-        con.power_on()
+        # We need to do a sleep here because we don't know how long it will
+        # take for a hard_reset to be effected.
+        time.sleep(30)
 
-        # Wait for node to come back up after reboot
-        misc.reconnect(None, 300, [self.client_remote])
+        try:
+            # Wait for node to come back up after reboot
+            misc.reconnect(None, 300, [self.client_remote])
+        except:
+            # attempt to get some useful debug output:
+            con = orchestra_remote.getRemoteConsole(self.client_remote.hostname,
+                                                    self.ipmi_user,
+                                                    self.ipmi_password,
+                                                    self.ipmi_domain)
+            con.check_status(timeout=60)
+            raise
+
+        # Remove mount directory
+        self.client_remote.run(args=['uptime'], timeout=10)
 
         # Remove mount directory
         self.client_remote.run(