Test OOM Kill on client when heal is in progress on 1*3 arbiter volume

Change-Id: I47d0ac4afac44442bd877243c45581df83c6a2e7
Signed-off-by: Vitalii Koriakov <vkoriako@redhat.com>

1 files changed, 201 insertions, 0 deletions

diff --git a/tests/functional/arbiter/test_oom_on_client_heal_is_in_progress_arbiter.py b/tests/functional/arbiter/test_oom_on_client_heal_is_in_progress_arbiter.py
new file mode 100755
index 000000000..26854729a
--- /dev/null
+++ b/tests/functional/arbiter/test_oom_on_client_heal_is_in_progress_arbiter.py
@@ -0,0 +1,201 @@
+#  Copyright (C) 2016-2018  Red Hat, Inc. <http://www.redhat.com>
+#
+#  This program is free software; you can redistribute it and/or modify
+#  it under the terms of the GNU General Public License as published by
+#  the Free Software Foundation; either version 2 of the License, or
+#  any later version.
+#
+#  This program is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+#  GNU General Public License for more details.
+#
+#  You should have received a copy of the GNU General Public License along
+#  with this program; if not, write to the Free Software Foundation, Inc.,
+#  51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+
+from glusto.core import Glusto as g
+from glustolibs.gluster.gluster_base_class import (GlusterBaseClass, runs_on)
+from glustolibs.gluster.exceptions import ExecutionError
+from glustolibs.gluster.brick_libs import (bring_bricks_offline,
+                                           bring_bricks_online,
+                                           are_bricks_offline,
+                                           get_all_bricks)
+from glustolibs.io.utils import (validate_io_procs,
+                                 list_all_files_and_dirs_mounts,
+                                 wait_for_io_to_complete)
+from glustolibs.gluster.lib_utils import list_files
+from glustolibs.misc.misc_libs import upload_scripts
+
+
+@runs_on([['replicated'],
+          ['glusterfs', 'nfs', 'cifs']])
+class ArbiterSelfHealTests(GlusterBaseClass):
+    """
+            Arbiter Self-Heal tests
+    """
+    @classmethod
+    def setUpClass(cls):
+        # Calling GlusterBaseClass setUpClass
+        GlusterBaseClass.setUpClass.im_func(cls)
+
+        # Upload io scripts for running IO on mounts
+        g.log.info("Upload io scripts to clients %s for running IO on mounts",
+                   cls.clients)
+        script_local_path = ("/usr/share/glustolibs/io/scripts/"
+                             "file_dir_ops.py")
+        cls.script_upload_path = ("/usr/share/glustolibs/io/scripts/"
+                                  "file_dir_ops.py")
+        ret = upload_scripts(cls.clients, [script_local_path])
+        if not ret:
+            raise ExecutionError("Failed to upload IO scripts to clients %s"
+                                 % cls.clients)
+        g.log.info("Successfully uploaded IO scripts to clients %s",
+                   cls.clients)
+
+    def setUp(self):
+        """
+        setUp method for every test
+        """
+        # calling GlusterBaseClass setUp
+        GlusterBaseClass.setUp.im_func(self)
+
+        self.all_mounts_procs = []
+        self.io_validation_complete = False
+
+        # Setup Volume and Mount Volume
+        g.log.info("Starting to Setup Volume %s", self.volname)
+        ret = self.setup_volume_and_mount_volume(self.mounts)
+        if not ret:
+            raise ExecutionError("Failed to Setup_Volume and Mount_Volume")
+        g.log.info("Successful in Setup Volume and Mount Volume")
+
+    def tearDown(self):
+        """
+        If test method failed before validating IO, tearDown waits for the
+        IO's to complete and checks for the IO exit status
+
+        Cleanup and umount volume
+        """
+        if not self.io_validation_complete:
+            g.log.info("Wait for IO to complete as IO validation did not "
+                       "succeed in test method")
+            ret = wait_for_io_to_complete(self.all_mounts_procs, self.mounts)
+            if not ret:
+                raise ExecutionError("IO failed on some of the clients")
+            g.log.info("IO is successful on all mounts")
+
+            # List all files and dirs created
+            g.log.info("List all files and directories:")
+            ret = list_all_files_and_dirs_mounts(self.mounts)
+            if not ret:
+                raise ExecutionError("Failed to list all files and dirs")
+            g.log.info("Listing all files and directories is successful")
+
+        # Cleanup and umount volume
+        g.log.info("Starting to Unmount Volume and Cleanup Volume")
+        ret = self.unmount_volume_and_cleanup_volume(mounts=self.mounts)
+        if not ret:
+            raise ExecutionError("Failed to umount the vol & cleanup Volume")
+        g.log.info("Successful in umounting the volume and Cleanup")
+
+        # Calling GlusterBaseClass teardown
+        GlusterBaseClass.tearDown.im_func(self)
+
+    def test_oom_on_client_heal_in_progress(self):
+        """
+        - Create a 1x(2+1) arbiter replicate volume
+        - Create IO
+        - Bring down the 1-st data brick while creating IO
+        - Bring up the 1-st data brick after creating and checking IO
+        - Bring down the 3-d arbiter brick
+        - Bring up the 3-d arbiter brick
+        - Check there no any oom by listing the files from mountpoint
+        """
+
+        # Creating IO on client side
+        for mount_obj in self.mounts:
+            g.log.info("Generating data for %s:%s",
+                       mount_obj.client_system, mount_obj.mountpoint)
+            # Create files
+            g.log.info('Creating files...')
+            command = ("python %s create_files "
+                       "-f 1000 "
+                       "--fixed-file-size 10k "
+                       "%s"
+                       % (self.script_upload_path, mount_obj.mountpoint))
+
+            proc = g.run_async(mount_obj.client_system, command,
+                               user=mount_obj.user)
+            self.all_mounts_procs.append(proc)
+        self.io_validation_complete = False
+
+        # get the bricks for the volume
+        g.log.info("Fetching bricks for the volume: %s", self.volname)
+        bricks_list = get_all_bricks(self.mnode, self.volname)
+        g.log.info("Brick list: %s", bricks_list)
+
+        # Bring brick 1 offline
+        bricks_to_bring_offline = [bricks_list[0]]
+        g.log.info('Bringing bricks %s offline...', bricks_to_bring_offline)
+        ret = bring_bricks_offline(self.volname, bricks_to_bring_offline)
+        self.assertTrue(ret, 'Failed to bring bricks %s offline' %
+                        bricks_to_bring_offline)
+
+        ret = are_bricks_offline(self.mnode, self.volname,
+                                 bricks_to_bring_offline)
+        self.assertTrue(ret, 'Bricks %s are not offline'
+                        % bricks_to_bring_offline)
+        g.log.info('Bringing bricks %s offline is successful',
+                   bricks_to_bring_offline)
+
+        # Validate IO
+        g.log.info("Wait for IO to complete and validate IO ...")
+        ret = validate_io_procs(self.all_mounts_procs, self.mounts)
+        self.assertTrue(ret, "IO failed on some of the clients")
+        self.io_validation_complete = True
+        g.log.info("IO is successful on all mounts")
+
+        # Bring 1-st brick online
+        g.log.info('Bringing bricks %s online...', bricks_to_bring_offline)
+        ret = bring_bricks_online(self.mnode, self.volname,
+                                  bricks_to_bring_offline)
+        self.assertTrue(ret, 'Failed to bring bricks %s online'
+                        % bricks_to_bring_offline)
+        g.log.info('Bringing bricks %s online is successful',
+                   bricks_to_bring_offline)
+
+        # Bring brick 3 offline
+        bricks_to_bring_offline = [bricks_list[-1]]
+        g.log.info('Bringing bricks %s offline...', bricks_to_bring_offline)
+        ret = bring_bricks_offline(self.volname, bricks_to_bring_offline)
+        self.assertTrue(ret, 'Failed to bring bricks %s offline' %
+                        bricks_to_bring_offline)
+
+        ret = are_bricks_offline(self.mnode, self.volname,
+                                 bricks_to_bring_offline)
+        self.assertTrue(ret, 'Bricks %s are not offline'
+                        % bricks_to_bring_offline)
+        g.log.info('Bringing bricks %s offline is successful',
+                   bricks_to_bring_offline)
+
+        # Bring brick 3 online
+        g.log.info('Bringing bricks %s online...', bricks_to_bring_offline)
+        ret = bring_bricks_online(self.mnode, self.volname,
+                                  bricks_to_bring_offline)
+        self.assertTrue(ret, 'Failed to bring bricks %s online'
+                        % bricks_to_bring_offline)
+        g.log.info('Bringing bricks %s online is successful',
+                   bricks_to_bring_offline)
+
+        # Get file list from mountpoint
+        g.log.info('Getting file list from mountpoints...')
+        for mount_obj in self.mounts:
+            g.log.info("Getting file list for %s:%s",
+                       mount_obj.client_system, mount_obj.mountpoint)
+            g.log.info('Getting file list...')
+            file_list = list_files(mount_obj.client_system,
+                                   mount_obj.mountpoint)
+            self.assertIsNotNone(file_list)
+        g.log.info('Getting file list from mountpoints finished successfully')