We shouldn't clean up class-scope resources(_cleanup). We should create local
ones(cleanup) then clean it.
So:
cls._cleanup would be used for class-scope resources.
self.cleanup would be used for each unit.
This result in countless test case failure because many test case units didn't
clean up after it.
The regression test result should be improved quite a bit after this change.
- No need to cleanup offerings as they are lightweight entities
- Cleanup order changed to clean accounts, networks and then the domains
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit c44fa2cc20)
Make the size explicit parameter to random_gen() for snapshot data
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 874d59847aa6bfd12e6967891a21f1746987b566)
since random_gen can take an id to prepend a testcase identifier to the
created cloud resource, we need to explicitly specify the size param. In
vmsnapshot test create the random_data string of size (100)
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 5bb3a56acac440b022ed6c45f97d0d85dcc94b96)
Now password is sent as clear text as per CLOUDSTACK-1734
So changed marvin to handle this. Plus domainid was not
passed in the testcase and marvin used "domainid" instead of
"domainId" as a parameter. Fixed these two errors.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit dba09791fc99c342974d8ca1c69609c04f1a5512)
- SourceHost is an id
- Invalid format in debug() message corrected
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit ce4dd723ab486894523f08f48b6054f614be4d85)
when an account is deleted listAccounts raises an exception.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit aa9bc4413efad4eccdf8e336f66ea8c8168fbbda)
Test VPC Release IP assigned to LB rule fails because we don't pass in
the apiclient to it.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 2f0a0a61363074ffa959d7ff9596a1b07b2de40a)
During switchover of master->backup->master, the rvr goes through
UNKNOWN, FAULT, MASTER/BACKUP. In case of improper transition the state
is FAULT. Include additional state check in the tests.
(cherry picked from commit 0c699394f246bc2b052a22d88b3f938ba8a72770)
The guestcidr is no longer specified directly. Use the netmask and
gateway combination to create custom cidr network
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit a4d902fc9b48e963d5987965207e3745e1728741)
Adding some docstrings to the tests so the test runner reports them.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 296c2f6a70)
We don't need to delete the volume resource as part of cleanup as the
storage gc is expected to cleanup orphaned volumes in deleted
accounts/projects.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 8bc3757e80)
API call listprojectaccounts failed, and test case
test_projects.py:test_08_cleanup_after_project_delete failed due to
this
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit 52fa8532bb)
Emails belong to user's within accounts. Refer to the first user within
the account and use his email for account invites.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
The network GC test is a long running test as it waits for
network.gc.wait + network.gc.interval * 2 to verify that the network
cleanups happen corectly. This test is also part of the regression suite
and is redundant in the smoke. The test run takes 30m longer because of
this test included in smoke.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
After you scale the vm, user is expected (oddly) to update the virtual
machine to signal that the PV tools are installed. This tells cloudstack
that the VM has been successfully scaled up/down.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
(cherry picked from commit c8ad4e5ced)
updateVirutalMachine fails if no args are passed. this test also further
verifies that the displayname is changed when vm is started up
successfully.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
User Dispersion is a best effort of deployment planning and cannot
always guarantee the dispersion logic will work. However this does not
fail the deployment. Adjust the test so as to print a debug and exit
instead of fail.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
the nosetests runner will not discover tests unless the tests start with
the test_ prefix. When wrapping by the log_test_exceptions method the
name of the method changes and is not test_ anymore causing the runner
to skip the test. Included the __doc__ string as well so the runner can
report results correctly.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
CLOUDSTACK-2457: No check for input values,special character,-ive values can be assigne to GLobal parameter "cluster.cpu.allocated.capacity.disablethreshold"
Added validation for float values and checking limit for weight based parameters by creating a set of parameters.
Signed off by : nitin mehta<nitin.mehta@citrix.com>
Sub domain removed before parent domain. Also fixed indentation issue in
cleanup list of class TestServiceOfferingHierarchy
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
1.Current Implementation assumes 6 as the max volumes that can be
attached to disk. Actually it depends on the hypervisor capabilities.
2.Modified the script to get the max limit from hypervsior capabilities.
Signed-off-by: sanjeevneelarapu <sanjeev.neelarapu@citrix.com>
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Do not wait until gc finishes cleanup of resources. This delays tests
and is dependant on the deployment. Some have large cleanup.intervals.
Assume cleanup succeeds and proceed
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
CIDR dictionary changes in the test causing the subsequent test to fail.
Reinitialize the dictionary at the beginning of the test
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Each module of redundant router contains related tests for
services, network rules, upgrades, redundancy etc.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Cleanups need not wait at teardown for accounts to be removed. This
slows down the test case. Send the cleanup request and forget instead
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Resources are ordered so that the cleanup succeeds.
domains are deleted after accounts under the domain are removed.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
On kvm the disk device serial order is /dev/vda|vdb|vdc and so on.
This also fixes CLOUDSTACK-3018 and removes the redundant test in
blocker bugs. The snapshot related tests are now include in snapshots
suite.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
user.concentrated pod planner will try and put the VMs in the same pod.
Correcting the assertion to verify same pod and not same cluster. Test
may have passed earlier because of presence of single pod in the
deployment.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Many vpc tests fail because of incorrect apiClient passed in to create
the VPC, network etc. The exact method used is getUserApiClient to fetch
the apiclient for a specific user.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
All tests that could possible disrupt the runs of other tests because of
putting resources into maintenace will be put under maint/. This should
allow us to run the tests sequentially when other tests are not running
on a deployment.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
If you do not give a name during vm creation, UUID will be set for Name
and Display name will be blank.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
When the router for an account is deleted listRouters throws an
exception, handle the exception to pass the test.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Bug in listing hosts that can be used for deploying VMs. Use Routing
hosts to identify hosts regardless of hypervisor_type.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Concentration or Dispersion granularity is at the Cluster level and not
at the host level. So correcting the test to ensure
a. concentrated planner puts the VMs in same cluster
b. dispersed planner puts the Vms in diff't clusters
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
- Removing redundant creation of VPC offerings
- Removing cleanup based on configurations.GC should happen by default
- Speed up the run by not waiting for complete gc.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
NetworkOfferings now require a mandatory scheme in the
serviceCapabilityList to create a VPC loadbalancer on the public side.
This commit fixes the test for VPC networks. Additionally there needs to
be a fix for making this the default behaviour so as not to hurt the
backwards compatibility.
test still fails because of CLOUDSTACK-2915 however which is a related
network ACL backwards compat issue. See bug for more details.
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Skip the tests if the VM deploys on a VmWare host since we do not
support resizing volumes on vmware (yet)
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
Refactoring the internal lb test to use the marvin libraries. Also added
reqd methods for the internal lb related APIs into integration.lib
Signed-off-by: Prasanna Santhanam <tsp@apache.org>
This feature allows a user to deploy VMs only in the resources dedicated to his account or domain.
1. Resources(Zones, Pods, Clusters or hosts) can be dedicated to an account or domain.
Implemented 12 new APIs to dedicate/list/release resources:
- dedicateZone, listDedicatedZones, releaseDedicatedZone for a Zone.
- dedicatePod, listDedicatedPods, releaseDedicatedPod for a Pod.
- dedicateCluster, listDedicatedClusters, releaseDedicatedCluster for a Cluster
- dedicateHost, listDedicatedHosts, releaseDedicatedHost for a Host.
2. Once a resource(eg. pod) is dedicated to an account, other resources(eg. clusters/hosts) inside that cannot be further dedicated.
3. Once a resource is dedicated to a domain, other resources inside that can be further dedicated to its sub-domain or account.
4. If any resource (eg.cluster) is dedicated to a account/domain, then resources(eg. Pod) above that cannot be dedicated to different accounts/domain (not belonging to the same domain)
5. To use Explicit dedication, user needs to create an Affinity Group of type 'ExplicitDedication'
6. A VM can be deployed with the above affinity group parameter as an input.
7. A new ExplicitDedicationProcessor has been added which will process the affinity group of type 'Explicit Dedication' for a deployment of a VM that demands dedicated resources.
This processor implements the AffinityGroupProcessor adapter. This processor will update the avoid list.
8. A VM requesting dedication will be deployed on dedicatd resources if available with the user account.
9. A VM requesting dedication can also be deployed on the dedicated resources available with the parent domains iff no dedicated resources are available with the current user's account or
domain.
10. A VM (without dedication) can be deployed on shared host but not on dedicated hosts.
11. To modify the dedication, the resource has to be released first.
12. Existing Private zone functionality has been redirected to Explicit dedication of zones.
13. Updated the db upgrade schema script. A new table "dedicated_resources" has been added.
14. Added the right permissions in commands.properties
15. Unit tests: For the new APIs and Service, added unit tests under : plugins/dedicated-resources/test/org/apache/cloudstack/dedicated/DedicatedApiUnitTest.java
16. Marvin Test: To dedicate host, create affinity group, deploy-vm, check if vm is deployed on the dedicated host.
Sowmya Krishnan
Prasanna Santhanam
Likitha Shetty
Sheng Yang
Sanjay Tripathi
Girish Shilamkar
Gaurav Aradhye
sailajam
ashutoshkelkar
SrikanteswaraRao Talluri
sanjeevneelarapu
rayeesn
suresh sadhu
Harikrishna Patnala
Edison Su
Wei Zhou
Edison Su
Min Chen
Saksham Srivastava
Anthony Xu