Add vlan/macvlan/ipvlan incontainer master tests #27700
Conversation
mmirecki
force-pushed
the
incontainer_vlan
branch
from
c27974c
to
b000e76
Compare
mmirecki
force-pushed
the
incontainer_vlan
branch
from
b000e76
to
ddad1d8
Compare
mmirecki
force-pushed
the
incontainer_vlan
branch
from
ddad1d8
to
dd70407
Compare
|
/retest-required |
mmirecki
force-pushed
the
incontainer_vlan
branch
2 times, most recently
from
dc34afc
to
a928a3e
Compare
test/extended/networking/vlan.go
Outdated
| pod.Spec.Containers[0].Command = []string{"/bin/bash", "-c", fmt.Sprintf("ping -c 1 %s", ip1)} | ||
| }) | ||
|
|
||
| g.By("having another pod on different vlan unanble to ping the first pod") |
There was a problem hiding this comment.
| g.By("having another pod on different vlan unanble to ping the first pod") | |
| g.By("having another pod on different vlan unable to ping the first pod") |
There was a problem hiding this comment.
@mmirecki looks like this nit got missed.
|
/retest-required |
test/extended/networking/vlan.go
Outdated
| if err != nil { | ||
| return false, err | ||
| } | ||
| return retrievedPod.Status.ContainerStatuses[0].Ready == false, nil |
There was a problem hiding this comment.
STEP: having another pod on different vlan unanble to ping the first pod 02/20/23 22:04:26.144
E0220 22:04:26.297917 58812 runtime.go:79] Observed a panic: runtime.boundsError{x:0, y:0, signed:true, code:0x0} (runtime error: index out of range [0] with length 0)
goroutine 766 [running]:
k8s.io/apimachinery/pkg/util/runtime.logPanic({0x825abe0?, 0xc0048cf980})
k8s.io/apimachinery@v0.25.0/pkg/util/runtime/runtime.go:75 +0x99
k8s.io/apimachinery/pkg/util/runtime.HandleCrash({0x0, 0x0, 0x0?})
k8s.io/apimachinery@v0.25.0/pkg/util/runtime/runtime.go:49 +0x75
panic({0x825abe0, 0xc0048cf980})
runtime/panic.go:884 +0x212
github.com/openshift/origin/test/extended/networking.glob..func14.1.3()
github.com/openshift/origin/test/extended/networking/vlan.go:129 +0xd5
k8s.io/apimachinery/pkg/util/wait.ConditionFunc.WithContext.func1({0x18, 0xc0000d6c00})
k8s.io/apimachinery@v0.25.0/pkg/util/wait/wait.go:222 +0x1b
k8s.io/apimachinery/pkg/util/wait.runConditionWithCrashProtectionWithContext({0x9462ed8?, 0xc00007e130?}, 0xc004803c28?)
k8s.io/apimachinery@v0.25.0/pkg/util/wait/wait.go:235 +0x57
k8s.io/apimachinery/pkg/util/wait.poll({0x9462ed8, 0xc00007e130}, 0x68?,
There was a problem hiding this comment.
The test probably executes before the container is created.
Adding a check for len of containers.
mmirecki
force-pushed
the
incontainer_vlan
branch
from
a928a3e
to
ab5d814
Compare
Signed-off-by: mmirecki <mmirecki@redhat.com>
mlguerrero12
force-pushed
the
incontainer_vlan
branch
from
ab5d814
to
5c0c9a4
Compare
|
/approve |
openshift-ci
bot
added
the
approved
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: bparees, cgoncalves, mlguerrero12, mmirecki The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
@mmirecki: The following tests failed, say
Full PR test history. Your PR dashboard. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
| var _ = g.Describe("[sig-network][Feature:vlan]", func() { | ||
| oc := exutil.NewCLI("vlan") | ||
| f := oc.KubeFramework() | ||
| for _, pluginType := range []string{"vlan", "ipvlan", "macvlan"} { |
There was a problem hiding this comment.
@mmirecki You can still use a for loop for the acutal testcase logic, but don't for-loop the setup. Do that with BeforeEach()/AfterEach() so that things get cleaned up correctly too.
Anything that's common between the tests can go into the BeforeEach() and be a variable declared inside the top-level Describe(). BeforeEach() should also generate a unique bridge name to pass to the CNI plugin. Be a bit careful here of parallel tests since each spec can run in parallel and variables sometimes get evaluated before the test is actually run; we can handle that in the resubmit PR review though.
AfterEach() needs to clean up anything the test creates in the kube API, like all the attachment definitions if they exist, so that errors that stop the test in the middle don't leave them around.
| podName1, podName2, podName3 := "pod1", "pod2", "pod3" | ||
| vlan1, vlan2, otherVlan := "vlan01", "vlan02", "vlan03" | ||
| ip1, ip2, otherIp := "10.10.0.2", "10.10.0.3", "10.10.0.4" | ||
| bridge := "br0" |
There was a problem hiding this comment.
Per above, this needs to be auto-generated as a unique name in the BeforeEach() so that each test uses a different name.
Tests for the following PRs:
containernetworking/plugins#781
containernetworking/plugins#813
containernetworking/plugins#814
Signed-off-by: mmirecki mmirecki@redhat.com