Re: [Qemu-devel] [PATCH v9 4/4] qemu-iotests: Add TestRepairQuorum to 041 to test drive-mirror node-name mode.

[Benoît Canet]

The Saturday 14 Jun 2014 à 02:44:29 (+0200), Max Reitz wrote :
> On 11.06.2014 15:24, Benoît Canet wrote:
> >The to-replace-node-name is designed to allow repairing of broken Quorum 
> >file.
> 
> "a broken Quorum file" or "broken Quorum files".
> 
> >This patch introduce a new class TestRepairQuorum testing that the feature
> 
> *introduces
> 
> >works.
> >Some further work will be done on QEMU to improve the robutness of the tests.
> 
> *robustness
> 
> >Signed-off-by: Benoit Canet <address@hidden>
> >---
> >  tests/qemu-iotests/041     | 196 
> > ++++++++++++++++++++++++++++++++++++++++++++-
> >  tests/qemu-iotests/041.out |   4 +-
> >  2 files changed, 194 insertions(+), 6 deletions(-)
> >
> >diff --git a/tests/qemu-iotests/041 b/tests/qemu-iotests/041
> >index ec470b2..c0480e1 100755
> >--- a/tests/qemu-iotests/041
> >+++ b/tests/qemu-iotests/041
> >@@ -28,6 +28,12 @@ target_backing_img = os.path.join(iotests.test_dir, 
> >'target-backing.img')
> >  test_img = os.path.join(iotests.test_dir, 'test.img')
> >  target_img = os.path.join(iotests.test_dir, 'target.img')
> >+quorum_img1 = os.path.join(iotests.test_dir, 'quorum1.img')
> >+quorum_img2 = os.path.join(iotests.test_dir, 'quorum2.img')
> >+quorum_img3 = os.path.join(iotests.test_dir, 'quorum3.img')
> >+quorum_repair_img = os.path.join(iotests.test_dir, 'quorum_repair.img')
> >+quorum_snapshot_file = os.path.join(iotests.test_dir, 'quorum_snapshot.img')
> >+
> >  class ImageMirroringTestCase(iotests.QMPTestCase):
> >      '''Abstract base class for image mirroring test cases'''
> >@@ -42,8 +48,8 @@ class ImageMirroringTestCase(iotests.QMPTestCase):
> >                      ready = True
> >      def wait_ready_and_cancel(self, drive='drive0'):
> >-        self.wait_ready(drive)
> >-        event = self.cancel_and_wait()
> >+        self.wait_ready(drive=drive)
> >+        event = self.cancel_and_wait(drive=drive)
> 
> My Python is very bad (if existent at all), but is this
> "drive=drive" really necessary? Wouldn't simply "drive" suffice?

drive=drive means pass the drive variable in the parameter of the callee named
drive.
cancel_and_wait have multiple named parameters I need to pass the variable into
the right one.

> 
> >          self.assertEquals(event['event'], 'BLOCK_JOB_COMPLETED')
> >          self.assert_qmp(event, 'data/type', 'mirror')
> >          self.assert_qmp(event, 'data/offset', self.image_len)
> >@@ -52,12 +58,12 @@ class ImageMirroringTestCase(iotests.QMPTestCase):
> >      def complete_and_wait(self, drive='drive0', wait_ready=True):
> >          '''Complete a block job and wait for it to finish'''
> >          if wait_ready:
> >-            self.wait_ready()
> >+            self.wait_ready(drive=drive)
> >          result = self.vm.qmp('block-job-complete', device=drive)
> >          self.assert_qmp(result, 'return', {})
> >-        event = self.wait_until_completed()
> >+        event = self.wait_until_completed(drive=drive)
> >          self.assert_qmp(event, 'data/type', 'mirror')
> >  class TestSingleDrive(ImageMirroringTestCase):
> >@@ -718,5 +724,187 @@ class TestUnbackedSource(ImageMirroringTestCase):
> >          self.complete_and_wait()
> >          self.assert_no_active_block_jobs()
> >+class TestRepairQuorum(ImageMirroringTestCase):
> >+    """ This class test quorum file repair using drive-mirror.
> >+        It's mostly a fork of TestSingleDrive """
> >+    image_len = 1 * 1024 * 1024 # MB
> >+    IMAGES = [ quorum_img1, quorum_img2, quorum_img3 ]
> >+
> >+    def setUp(self):
> >+        self.vm = iotests.VM()
> >+
> >+        # Add each individual quorum images
> >+        for i in self.IMAGES:
> >+            qemu_img('create', '-f', iotests.imgfmt, i,
> >+                     str(TestSingleDrive.image_len))
> >+            # Assign a node name to each quorum image in order to manipulate
> >+            # them
> >+            opts = "node-name=img%i" % self.IMAGES.index(i)
> >+            self.vm = self.vm.add_drive(i, opts)
> 
> Just an idea: As you are constantly swapping out img1, why don't you
> write some data to img2 and img3 and different data to img1 to test
> it all how it's supposed to work?

Hmm we have another test for quorum file corruption repairing.
This one is for major failure like the lost of a filer containing a particular
quorum file.

> 
> >+
> >+        self.vm.launch()
> >+
> >+        #assemble the quorum block device from the individual files
> >+        args = { "options" : { "driver": "quorum", "id": "quorum0",
> >+                 "vote-threshold": 2, "children": [ "img0", "img1", "img2" 
> >] } }
> >+        result = self.vm.qmp("blockdev-add", **args)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+
> >+    def tearDown(self):
> >+        self.vm.shutdown()
> >+        for i in self.IMAGES + [ quorum_repair_img ]:
> >+            # Do a try/except because the test may have deleted some images
> >+            try:
> >+                os.remove(i)
> >+            except OSError:
> >+                pass
> >+
> >+    def test_complete(self):
> >+        self.assert_no_active_block_jobs()
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name="repair0",
> >+                             replaces="img1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        self.complete_and_wait(drive="quorum0")
> >+        result = self.vm.qmp('query-named-block-nodes')
> >+        self.assert_qmp(result, 'return[0]/file', quorum_repair_img)
> 
> Hm, why do you know this is at index 0?

Because I ran the comand manually and it's predictible. It's all linked list
insertion and traversal.

> 
> >+        # TODO: a better test requiring some QEMU infrastructure will be 
> >added
> >+        #       to check that this file is really driven by quorum
> >+        self.vm.shutdown()
> >+        self.assertTrue(iotests.compare_images(quorum_img2, 
> >quorum_repair_img),
> >+                        'target image does not match source after 
> >mirroring')
> >+
> >+    def test_cancel(self):
> >+        self.assert_no_active_block_jobs()
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name="repair0",
> >+                             replaces="img1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        self.cancel_and_wait(drive="quorum0", force=True)
> >+        # here we check that the last registered quorum file has not been
> >+        # swapped out and unref
> >+        result = self.vm.qmp('query-named-block-nodes')
> >+        self.assert_qmp(result, 'return[0]/file', quorum_img3)
> 
> Why would img3 be affected at all? Aren't you trying to replace
> img1? (Also, again, why index 0?)

It means quorum_img3 should be at position 0 in the linked list: the last 
inserted
quorum file since we canceled the drive-mirror.

> 
> >+        self.vm.shutdown()
> >+
> >+    def test_cancel_after_ready(self):
> >+        self.assert_no_active_block_jobs()
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name="repair0",
> >+                             replaces="img1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        self.wait_ready_and_cancel(drive="quorum0")
> >+        result = self.vm.qmp('query-named-block-nodes')
> >+        # here we check that the last registered quorum file has not been
> >+        # swapped out and unref
> >+        self.assert_qmp(result, 'return[0]/file', quorum_img3)
> 
> Same here.
same

> 
> >+        self.vm.shutdown()
> >+        self.assertTrue(iotests.compare_images(quorum_img2, 
> >quorum_repair_img),
> >+                        'target image does not match source after 
> >mirroring')
> >+
> >+    def test_pause(self):
> >+        self.assert_no_active_block_jobs()
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name="repair0",
> >+                             replaces="img1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        result = self.vm.qmp('block-job-pause', device='quorum0')
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        time.sleep(1)
> >+        result = self.vm.qmp('query-block-jobs')
> >+        offset = self.dictpath(result, 'return[0]/offset')
> >+
> >+        time.sleep(1)
> >+        result = self.vm.qmp('query-block-jobs')
> >+        self.assert_qmp(result, 'return[0]/offset', offset)
> >+
> >+        result = self.vm.qmp('block-job-resume', device='quorum0')
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        self.complete_and_wait(drive="quorum0")
> >+        self.vm.shutdown()
> >+        self.assertTrue(iotests.compare_images(quorum_img2, 
> >quorum_repair_img),
> >+                        'target image does not match source after 
> >mirroring')
> >+
> >+    def test_medium_not_found(self):
> >+        result = self.vm.qmp('drive-mirror', device='ide1-cd0', sync='full',
> 
> I don't know whether this device is supposed to exist forever, but
> if it disappears from default configuration one day, TestSingleDrive
> will break as well, so it's probably fine.
> 
> Max
> 
> >+                             node_name='repair0',
> >+                             replaces='img1',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+    def test_image_not_found(self):
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name='repair0',
> >+                             replaces='img1',
> >+                             mode='existing',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+    def test_device_not_found(self):
> >+        result = self.vm.qmp('drive-mirror', device='nonexistent', 
> >sync='full',
> >+                             node_name='repair0',
> >+                             replaces='img1',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
> >+
> >+    def test_wrong_sync_mode(self):
> >+        result = self.vm.qmp('drive-mirror', device='quorum0',
> >+                             node_name='repair0',
> >+                             replaces='img1',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+    def test_no_node_name(self):
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             replaces='img1',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+    def test_unexistant_replaces(self):
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name='repair0',
> >+                             replaces='img77',
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+    def test_after_a_quorum_snapshot(self):
> >+        result = self.vm.qmp('blockdev-snapshot-sync', node_name='img1',
> >+                             snapshot_file=quorum_snapshot_file,
> >+                             snapshot_node_name="snap1");
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name='repair0',
> >+                             replaces="img1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'error/class', 'GenericError')
> >+
> >+        result = self.vm.qmp('drive-mirror', device='quorum0', sync='full',
> >+                             node_name='repair0',
> >+                             replaces="snap1",
> >+                             target=quorum_repair_img, 
> >format=iotests.imgfmt)
> >+        self.assert_qmp(result, 'return', {})
> >+
> >+        self.complete_and_wait(drive="quorum0")
> >+        result = self.vm.qmp('query-named-block-nodes')
> >+        self.assert_qmp(result, 'return[0]/file', quorum_repair_img)
> >+        # TODO: a better test requiring some QEMU infrastructure will be 
> >added
> >+        #       to check that this file is really driven by quorum
> >+        self.vm.shutdown()
> >+
> >  if __name__ == '__main__':
> >      iotests.main(supported_fmts=['qcow2', 'qed'])
> >diff --git a/tests/qemu-iotests/041.out b/tests/qemu-iotests/041.out
> >index 6d9bee1..73e375a 100644
> >--- a/tests/qemu-iotests/041.out
> >+++ b/tests/qemu-iotests/041.out
> >@@ -1,5 +1,5 @@
> >-...........................
> >+......................................
> >  ----------------------------------------------------------------------
> >-Ran 27 tests
> >+Ran 38 tests
> >  OK
>