From 615d7886ee4d8cd6b9b9ab21e4e2b486fb4bb2fd Mon Sep 17 00:00:00 2001 From: root Date: Fri, 29 Dec 2017 12:10:53 +0530 Subject: TestCase Testing VolumeType change from replicated to Arbiter volume along with volume operations add-brick, remove-brick, replace-brick post volume type change Change-Id: I44a1ff6fab3228736ae9c83fe67b16c2e8c40adc Signed-off-by: Karan Sandha Signed-off-by: Vitalii Koriakov --- .../arbiter/brick_cases/test_brickcases.py | 394 +++++++++++++++++++++ 1 file changed, 394 insertions(+) create mode 100755 tests/functional/arbiter/brick_cases/test_brickcases.py (limited to 'tests/functional') diff --git a/tests/functional/arbiter/brick_cases/test_brickcases.py b/tests/functional/arbiter/brick_cases/test_brickcases.py new file mode 100755 index 000000000..5d3a03816 --- /dev/null +++ b/tests/functional/arbiter/brick_cases/test_brickcases.py @@ -0,0 +1,394 @@ +# Copyright (C) 2017-2018 Red Hat, Inc. +# +# 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. + +""" Test Arbiter Specific Cases""" + +from glusto.core import Glusto as g +from glustolibs.gluster.gluster_base_class import (GlusterBaseClass, runs_on) +from glustolibs.gluster.volume_libs import ( + log_volume_info_and_status, replace_brick_from_volume, + expand_volume, wait_for_volume_process_to_be_online, + verify_all_process_of_volume_are_online, shrink_volume) +from glustolibs.gluster.rebalance_ops import ( + rebalance_start, rebalance_status, wait_for_rebalance_to_complete) +from glustolibs.gluster.heal_libs import monitor_heal_completion +from glustolibs.gluster.exceptions import ExecutionError +from glustolibs.io.utils import (validate_io_procs, + list_all_files_and_dirs_mounts, + wait_for_io_to_complete) +from glustolibs.misc.misc_libs import upload_scripts + + +@runs_on([['replicated', 'distributed-replicated'], + ['glusterfs', 'cifs', 'nfs']]) +class GlusterArbiterVolumeTypeChangeClass(GlusterBaseClass): + """Class for testing Volume Type Change from replicated to + Arbitered volume + """ + @classmethod + def setUpClass(cls): + # Calling GlusterBaseClass setUpClass + GlusterBaseClass.setUpClass.im_func(cls) + + # Overriding the volume type to specifically test the volume type + # change from replicated to arbiter + if cls.volume_type == "replicated": + cls.volume['voltype'] = { + 'type': 'replicated', + 'replica_count': 2, + 'dist_count': 1, + 'transport': 'tcp'} + + if cls.volume_type == "distributed-replicated": + cls.volume['voltype'] = { + 'type': 'distributed-replicated', + 'dist_count': 2, + 'replica_count': 2, + 'transport': 'tcp'} + + # 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) + + cls.counter = 1 + # int: Value of counter is used for dirname-start-num argument for + # file_dir_ops.py create_deep_dirs_with_files. + + # The --dir-length argument value for file_dir_ops.py + # create_deep_dirs_with_files is set to 10 (refer to the cmd in setUp + # method). This means every mount will create + # 10 top level dirs. For every mountpoint/testcase to create new set of + # dirs, we are incrementing the counter by --dir-length value i.e 10 + # in this test suite. + + # If we are changing the --dir-length to new value, ensure the counter + # is also incremented by same value to create new set of files/dirs. + + def setUp(self): + """ + - Setup Volume and Mount Volume + """ + # Calling GlusterBaseClass setUpClass + GlusterBaseClass.setUpClass.im_func(self) + + self.all_mounts_procs = [] + self.io_validation_complete = False + + # Setup Volume and Mount Volume + g.log.info("Starting to Setup Volume and Mount Volume") + ret = self.setup_volume_and_mount_volume(mounts=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 + + Unmount Volume and Cleanup Volume + """ + # Wait for IO to complete if io validation is not executed in the + # test method + 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") + + # Unmount Volume and Cleanup 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 Unmount Volume and Cleanup Volume") + g.log.info("Successful in Unmount Volume and Cleanup Volume") + + # Calling GlusterBaseClass tearDown + GlusterBaseClass.tearDown.im_func(self) + + def test_replicated_to_arbiter_volume_change_with_volume_ops(self): + """ + - Change the volume type from replicated to arbiter + - Perform add-brick, rebalance on arbitered volume + - Perform replace-brick on arbitered volume + - Perform remove-brick on arbitered volume + """ + # pylint: disable=too-many-statements + + # Start IO on mounts + g.log.info("Starting IO on all mounts...") + self.all_mounts_procs = [] + for mount_obj in self.mounts: + g.log.info("Starting IO on %s:%s", mount_obj.client_system, + mount_obj.mountpoint) + cmd = ("python %s create_deep_dirs_with_files " + "--dirname-start-num %d " + "--dir-depth 2 " + "--dir-length 15 " + "--max-num-of-dirs 5 " + "--num-of-files 5 %s" % (self.script_upload_path, + self.counter, + mount_obj.mountpoint)) + proc = g.run_async(mount_obj.client_system, cmd, + user=mount_obj.user) + self.all_mounts_procs.append(proc) + self.counter = self.counter + 10 + self.io_validation_complete = False + + # Validate IO + g.log.info("Wait for IO to complete and validate IO ...") + ret = validate_io_procs(self.all_mounts_procs, self.mounts) + self.io_validation_complete = True + self.assertTrue(ret, "IO failed on some of the clients") + g.log.info("IO is successful on all mounts") + + # Adding bricks to make an Arbiter Volume + g.log.info("Adding bricks to convert to Arbiter Volume") + ret = expand_volume(self.mnode, self.volname, self.servers[2:], + self.all_servers_info, replica_count=1, + arbiter_count=1) + self.assertTrue(ret, ("Failed to expand the volume %s", self.volname)) + g.log.info("Changing volume to arbiter volume is successfull %s", + self.volname) + + # Log Volume Info and Status after changing the volume type from + # replicated to arbitered + g.log.info("Logging volume info and Status after changing to " + "arbitered volume") + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Wait for volume processes to be online + g.log.info("Wait for volume processes to be online") + ret = wait_for_volume_process_to_be_online(self.mnode, self.volname) + self.assertTrue(ret, ("Failed to wait for volume %s processes to " + "be online", self.volname)) + g.log.info("Successful in waiting for volume %s processes to be " + "online", self.volname) + + # Verifying all bricks online + g.log.info("Verifying volume's all process are online") + ret = verify_all_process_of_volume_are_online(self.mnode, self.volname) + self.assertTrue(ret, ("Volume %s : All process are not online", + self.volname)) + g.log.info("Volume %s : All process are online", self.volname) + + # Checking for heals to finish after changing the volume type + # from replicated to arbitered volume + g.log.info("Wait for self-heal to complete after changing the " + "volume type from replicated to arbitered volume") + ret = monitor_heal_completion(self.mnode, self.volname) + self.assertTrue(ret, ("Self heal didn't complete even after waiting " + "for 20 minutes.")) + g.log.info("self-heal is successful after changing the volume type " + "from replicated to arbitered volume") + + # Start IO on mounts + g.log.info("Starting IO on all mounts...") + self.all_mounts_procs = [] + for mount_obj in self.mounts: + g.log.info("Starting IO on %s:%s", mount_obj.client_system, + mount_obj.mountpoint) + cmd = ("python %s create_deep_dirs_with_files " + "--dirname-start-num %d " + "--dir-depth 2 " + "--dir-length 35 " + "--max-num-of-dirs 5 " + "--num-of-files 5 %s" % (self.script_upload_path, + self.counter, + mount_obj.mountpoint)) + proc = g.run_async(mount_obj.client_system, cmd, + user=mount_obj.user) + self.all_mounts_procs.append(proc) + self.counter = self.counter + 10 + self.io_validation_complete = False + + # Log Volume Info and Status before expanding the volume. + g.log.info("Logging volume info and Status before expanding volume") + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Start add-brick (subvolume-increase) + g.log.info("Start adding bricks to volume when IO in progress") + ret = expand_volume(self.mnode, self.volname, self.servers, + self.all_servers_info) + self.assertTrue(ret, ("Failed to expand the volume when IO in " + "progress on volume %s", self.volname)) + g.log.info("Expanding volume when IO in progress is successful on " + "volume %s", self.volname) + + # Log Volume Info and Status after expanding the volume + g.log.info("Logging volume info and Status after expanding volume") + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Wait for volume processes to be online + g.log.info("Wait for volume processes to be online") + ret = wait_for_volume_process_to_be_online(self.mnode, self.volname) + self.assertTrue(ret, ("Failed to wait for volume %s processes to " + "be online", self.volname)) + g.log.info("Successful in waiting for volume %s processes to be " + "online", self.volname) + + # Verify volume's all process are online + g.log.info("Verifying volume's all process are online") + ret = verify_all_process_of_volume_are_online(self.mnode, self.volname) + self.assertTrue(ret, ("Volume %s : All process are not online", + self.volname)) + g.log.info("Volume %s : All process are online", self.volname) + + # Start Rebalance + g.log.info("Starting Rebalance on the volume") + ret, _, _ = rebalance_start(self.mnode, self.volname) + self.assertEqual(ret, 0, ("Failed to start rebalance on the volume " + "%s", self.volname)) + g.log.info("Successfully started rebalance on the volume %s", + self.volname) + + # Log Rebalance status + g.log.info("Log Rebalance status") + _, _, _ = rebalance_status(self.mnode, self.volname) + + # Wait for rebalance to complete + g.log.info("Waiting for rebalance to complete") + ret = wait_for_rebalance_to_complete(self.mnode, self.volname) + self.assertTrue(ret, "Rebalance did not start " + "despite waiting for 5 mins") + g.log.info("Rebalance is successfully complete on the volume %s", + self.volname) + + # Log Volume Info and Status before replacing brick from the volume. + g.log.info("Logging volume info and Status before replacing brick " + "from the volume %s", self.volname) + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Replace brick from a sub-volume + g.log.info("Replace a faulty brick from the volume") + ret = replace_brick_from_volume(self.mnode, self.volname, + self.servers, self.all_servers_info) + self.assertTrue(ret, "Failed to replace faulty brick from the volume") + g.log.info("Successfully replaced faulty brick from the volume") + + # Log Volume Info and Status after replacing the brick + g.log.info("Logging volume info and Status after replacing brick " + "from the volume %s", self.volname) + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Wait for volume processes to be online + g.log.info("Wait for volume processes to be online") + ret = wait_for_volume_process_to_be_online(self.mnode, self.volname) + self.assertTrue(ret, ("Failed to wait for volume %s processes to " + "be online", self.volname)) + g.log.info("Successful in waiting for volume %s processes to be " + "online", self.volname) + + # Verify volume's all process are online + g.log.info("Verifying volume's all process are online") + ret = verify_all_process_of_volume_are_online(self.mnode, self.volname) + self.assertTrue(ret, ("Volume %s : All process are not online", + self.volname)) + g.log.info("Volume %s : All process are online", self.volname) + + # Wait for self-heal to complete + g.log.info("Wait for self-heal to complete") + ret = monitor_heal_completion(self.mnode, self.volname) + self.assertTrue(ret, "Self heal didn't complete even after waiting " + "for 20 minutes. 20 minutes is too much a time for " + "current test workload") + g.log.info("self-heal is successful after replace-brick operation") + + # Log Volume Info and Status before shrinking the volume. + g.log.info("Logging volume info and Status before shrinking volume") + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Shrinking volume by removing bricks from volume when IO in progress + g.log.info("Start removing bricks from volume when IO in progress") + ret = shrink_volume(self.mnode, self.volname) + self.assertTrue(ret, ("Failed to shrink the volume when IO in " + "progress on volume %s", self.volname)) + g.log.info("Shrinking volume when IO in progress is successful on " + "volume %s", self.volname) + + # Log Volume Info and Status after shrinking the volume + g.log.info("Logging volume info and Status after shrinking volume") + ret = log_volume_info_and_status(self.mnode, self.volname) + self.assertTrue(ret, ("Logging volume info and status failed on " + "volume %s", self.volname)) + g.log.info("Successful in logging volume info and status of volume %s", + self.volname) + + # Wait for volume processes to be online + g.log.info("Wait for volume processes to be online") + ret = wait_for_volume_process_to_be_online(self.mnode, self.volname) + self.assertTrue(ret, ("Failed to wait for volume %s processes to " + "be online", self.volname)) + g.log.info("Successful in waiting for volume %s processes to be " + "online", self.volname) + + # Verify volume's all process are online + g.log.info("Verifying volume's all process are online after " + "shrinking volume") + ret = verify_all_process_of_volume_are_online(self.mnode, self.volname) + self.assertTrue(ret, ("Volume %s : All process are not online", + self.volname)) + g.log.info("Volume %s : All process are online after shrinking volume", + self.volname) + + # 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") -- cgit