mirror of
https://github.com/containers/podman.git
synced 2025-12-12 09:50:25 +08:00
a071939893
Most Libpod containers are made via `pkg/specgen/generate` which includes code to generate an appropriate exit command which will handle unmounting the container's storage, cleaning up the container's network, etc. There is one notable exception: pod infra containers, which are made entirely within Libpod and do not touch pkg/specgen. As such, no cleanup process, network never cleaned up, bad things can happen. There is good news, though - it's not that difficult to add this, and it's done in this PR. Generally speaking, we don't allow passing options directly to the infra container at create time, but we do (optionally) proxy a pre-approved set of options into it when we create it. Add ExitCommand to these options, and set it at time of pod creation using the same code we use to generate exit commands for normal containers. Fixes #7103 Signed-off-by: Matthew Heon <mheon@redhat.com>
370 lines
11 KiB
Go
370 lines
11 KiB
Go
package test_bindings
|
|
|
|
import (
|
|
"net/http"
|
|
"strings"
|
|
"time"
|
|
|
|
"github.com/containers/podman/v2/libpod/define"
|
|
"github.com/containers/podman/v2/pkg/bindings"
|
|
"github.com/containers/podman/v2/pkg/bindings/pods"
|
|
"github.com/containers/podman/v2/pkg/specgen"
|
|
. "github.com/onsi/ginkgo"
|
|
. "github.com/onsi/gomega"
|
|
"github.com/onsi/gomega/gexec"
|
|
)
|
|
|
|
var _ = Describe("Podman pods", func() {
|
|
var (
|
|
bt *bindingTest
|
|
s *gexec.Session
|
|
newpod string
|
|
err error
|
|
)
|
|
|
|
BeforeEach(func() {
|
|
bt = newBindingTest()
|
|
newpod = "newpod"
|
|
bt.RestoreImagesFromCache()
|
|
bt.Podcreate(&newpod)
|
|
s = bt.startAPIService()
|
|
time.Sleep(1 * time.Second)
|
|
err := bt.NewConnection()
|
|
Expect(err).To(BeNil())
|
|
})
|
|
|
|
AfterEach(func() {
|
|
s.Kill()
|
|
bt.cleanup()
|
|
})
|
|
|
|
It("inspect pod", func() {
|
|
//Inspect an invalid pod name
|
|
_, err := pods.Inspect(bt.conn, "dummyname")
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ := bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusNotFound))
|
|
|
|
//Inspect an valid pod name
|
|
response, err := pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.Name).To(Equal(newpod))
|
|
})
|
|
|
|
// Test validates the list all api returns
|
|
It("list pod", func() {
|
|
//List all the pods in the current instance
|
|
podSummary, err := pods.List(bt.conn, nil)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(podSummary)).To(Equal(1))
|
|
|
|
// Start the pod
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
// Adding an alpine container to the existing pod
|
|
_, err = bt.RunTopContainer(nil, bindings.PTrue, &newpod)
|
|
Expect(err).To(BeNil())
|
|
podSummary, err = pods.List(bt.conn, nil)
|
|
// Verify no errors.
|
|
Expect(err).To(BeNil())
|
|
// Verify number of containers in the pod.
|
|
Expect(len(podSummary[0].Containers)).To(Equal(2))
|
|
|
|
// Add multiple pods and verify them by name and size.
|
|
var newpod2 string = "newpod2"
|
|
bt.Podcreate(&newpod2)
|
|
podSummary, err = pods.List(bt.conn, nil)
|
|
Expect(len(podSummary)).To(Equal(2))
|
|
var names []string
|
|
for _, i := range podSummary {
|
|
names = append(names, i.Name)
|
|
}
|
|
Expect(StringInSlice(newpod, names)).To(BeTrue())
|
|
Expect(StringInSlice("newpod2", names)).To(BeTrue())
|
|
})
|
|
|
|
// The test validates the list pod endpoint with passing filters as the params.
|
|
It("List pods with filters", func() {
|
|
newpod2 := "newpod2"
|
|
bt.Podcreate(&newpod2)
|
|
|
|
// Start the pod
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
_, err = bt.RunTopContainer(nil, bindings.PTrue, &newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
// Expected err with invalid filter params
|
|
filters := make(map[string][]string)
|
|
filters["dummy"] = []string{"dummy"}
|
|
filteredPods, err := pods.List(bt.conn, filters)
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ := bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusInternalServerError))
|
|
|
|
// Expected empty response with invalid filters
|
|
filters = make(map[string][]string)
|
|
filters["name"] = []string{"dummy"}
|
|
filteredPods, err = pods.List(bt.conn, filters)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(filteredPods)).To(BeNumerically("==", 0))
|
|
|
|
// Validate list pod with name filter
|
|
filters = make(map[string][]string)
|
|
filters["name"] = []string{newpod2}
|
|
filteredPods, err = pods.List(bt.conn, filters)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(filteredPods)).To(BeNumerically("==", 1))
|
|
var names []string
|
|
for _, i := range filteredPods {
|
|
names = append(names, i.Name)
|
|
}
|
|
Expect(StringInSlice("newpod2", names)).To(BeTrue())
|
|
|
|
// Validate list pod with id filter
|
|
filters = make(map[string][]string)
|
|
response, err := pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
id := response.ID
|
|
filters["id"] = []string{id}
|
|
filteredPods, err = pods.List(bt.conn, filters)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(filteredPods)).To(BeNumerically("==", 1))
|
|
names = names[:0]
|
|
for _, i := range filteredPods {
|
|
names = append(names, i.Name)
|
|
}
|
|
Expect(StringInSlice("newpod", names)).To(BeTrue())
|
|
|
|
// Using multiple filters
|
|
filters["name"] = []string{newpod}
|
|
filteredPods, err = pods.List(bt.conn, filters)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(filteredPods)).To(BeNumerically("==", 1))
|
|
names = names[:0]
|
|
for _, i := range filteredPods {
|
|
names = append(names, i.Name)
|
|
}
|
|
Expect(StringInSlice("newpod", names)).To(BeTrue())
|
|
})
|
|
|
|
// The test validates if the exists responds
|
|
It("exists pod", func() {
|
|
response, err := pods.Exists(bt.conn, "dummyName")
|
|
Expect(err).To(BeNil())
|
|
Expect(response).To(BeFalse())
|
|
|
|
// Should exit with no error and response should be true
|
|
response, err = pods.Exists(bt.conn, "newpod")
|
|
Expect(err).To(BeNil())
|
|
Expect(response).To(BeTrue())
|
|
})
|
|
|
|
// This test validates if All running containers within
|
|
// each specified pod are paused and unpaused
|
|
It("pause upause pod", func() {
|
|
// TODO fix this
|
|
Skip("Pod behavior is jacked right now.")
|
|
// Pause invalid container
|
|
_, err := pods.Pause(bt.conn, "dummyName")
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ := bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusNotFound))
|
|
|
|
// Adding an alpine container to the existing pod
|
|
_, err = bt.RunTopContainer(nil, bindings.PTrue, &newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
// Binding needs to be modified to inspect the pod state.
|
|
// Since we don't have a pod state we inspect the states of the containers within the pod.
|
|
// Pause a valid container
|
|
_, err = pods.Pause(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
response, err := pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStatePaused))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStatePaused))
|
|
}
|
|
|
|
// Unpause a valid container
|
|
_, err = pods.Unpause(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
response, err = pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStateRunning))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateRunning))
|
|
}
|
|
})
|
|
|
|
It("start stop restart pod", func() {
|
|
// Start an invalid pod
|
|
_, err = pods.Start(bt.conn, "dummyName")
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ := bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusNotFound))
|
|
|
|
// Stop an invalid pod
|
|
_, err = pods.Stop(bt.conn, "dummyName", nil)
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ = bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusNotFound))
|
|
|
|
// Restart an invalid pod
|
|
_, err = pods.Restart(bt.conn, "dummyName")
|
|
Expect(err).ToNot(BeNil())
|
|
code, _ = bindings.CheckResponseCode(err)
|
|
Expect(code).To(BeNumerically("==", http.StatusNotFound))
|
|
|
|
// Start a valid pod and inspect status of each container
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
response, err := pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStateRunning))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateRunning))
|
|
}
|
|
|
|
// Start an already running pod
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
|
|
// Stop the running pods
|
|
_, err = pods.Stop(bt.conn, newpod, nil)
|
|
Expect(err).To(BeNil())
|
|
response, _ = pods.Inspect(bt.conn, newpod)
|
|
Expect(response.State).To(Equal(define.PodStateExited))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateExited))
|
|
}
|
|
|
|
// Stop an already stopped pod
|
|
_, err = pods.Stop(bt.conn, newpod, nil)
|
|
Expect(err).To(BeNil())
|
|
|
|
_, err = pods.Restart(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
response, _ = pods.Inspect(bt.conn, newpod)
|
|
Expect(response.State).To(Equal(define.PodStateRunning))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateRunning))
|
|
}
|
|
})
|
|
|
|
// Test to validate all the pods in the stopped/exited state are pruned successfully.
|
|
It("prune pod", func() {
|
|
// Add a new pod
|
|
var newpod2 string = "newpod2"
|
|
bt.Podcreate(&newpod2)
|
|
// No pods pruned since no pod in exited state
|
|
pruneResponse, err := pods.Prune(bt.conn)
|
|
Expect(err).To(BeNil())
|
|
podSummary, err := pods.List(bt.conn, nil)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(podSummary)).To(Equal(2))
|
|
|
|
// Prune only one pod which is in exited state.
|
|
// Start then stop a pod.
|
|
// pod moves to exited state one pod should be pruned now.
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
_, err = pods.Stop(bt.conn, newpod, nil)
|
|
Expect(err).To(BeNil())
|
|
response, err := pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStateExited))
|
|
pruneResponse, err = pods.Prune(bt.conn)
|
|
Expect(err).To(BeNil())
|
|
// Validate status and record pod id of pod to be pruned
|
|
Expect(response.State).To(Equal(define.PodStateExited))
|
|
podID := response.ID
|
|
// Check if right pod was pruned
|
|
Expect(len(pruneResponse)).To(Equal(1))
|
|
Expect(pruneResponse[0].Id).To(Equal(podID))
|
|
// One pod is pruned hence only one pod should be active.
|
|
podSummary, err = pods.List(bt.conn, nil)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(podSummary)).To(Equal(1))
|
|
|
|
// Test prune multiple pods.
|
|
bt.Podcreate(&newpod)
|
|
_, err = pods.Start(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
_, err = pods.Start(bt.conn, newpod2)
|
|
Expect(err).To(BeNil())
|
|
_, err = pods.Stop(bt.conn, newpod, nil)
|
|
Expect(err).To(BeNil())
|
|
response, err = pods.Inspect(bt.conn, newpod)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStateExited))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateExited))
|
|
}
|
|
_, err = pods.Stop(bt.conn, newpod2, nil)
|
|
Expect(err).To(BeNil())
|
|
response, err = pods.Inspect(bt.conn, newpod2)
|
|
Expect(err).To(BeNil())
|
|
Expect(response.State).To(Equal(define.PodStateExited))
|
|
for _, i := range response.Containers {
|
|
Expect(define.StringToContainerStatus(i.State)).
|
|
To(Equal(define.ContainerStateExited))
|
|
}
|
|
_, err = pods.Prune(bt.conn)
|
|
Expect(err).To(BeNil())
|
|
podSummary, err = pods.List(bt.conn, nil)
|
|
Expect(err).To(BeNil())
|
|
Expect(len(podSummary)).To(Equal(0))
|
|
})
|
|
|
|
It("simple create pod", func() {
|
|
ps := specgen.PodSpecGenerator{}
|
|
ps.Name = "foobar"
|
|
_, err := pods.CreatePodFromSpec(bt.conn, &ps)
|
|
Expect(err).To(BeNil())
|
|
|
|
exists, err := pods.Exists(bt.conn, "foobar")
|
|
Expect(err).To(BeNil())
|
|
Expect(exists).To(BeTrue())
|
|
})
|
|
|
|
// Test validates the pod top bindings
|
|
It("pod top", func() {
|
|
var name string = "podA"
|
|
|
|
bt.Podcreate(&name)
|
|
_, err := pods.Start(bt.conn, name)
|
|
Expect(err).To(BeNil())
|
|
|
|
// By name
|
|
_, err = pods.Top(bt.conn, name, nil)
|
|
Expect(err).To(BeNil())
|
|
|
|
// With descriptors
|
|
output, err := pods.Top(bt.conn, name, []string{"user,pid,hpid"})
|
|
Expect(err).To(BeNil())
|
|
header := strings.Split(output[0], "\t")
|
|
for _, d := range []string{"USER", "PID", "HPID"} {
|
|
Expect(d).To(BeElementOf(header))
|
|
}
|
|
|
|
// With bogus ID
|
|
_, err = pods.Top(bt.conn, "IdoNotExist", nil)
|
|
Expect(err).ToNot(BeNil())
|
|
|
|
// With bogus descriptors
|
|
_, err = pods.Top(bt.conn, name, []string{"Me,Neither"})
|
|
Expect(err).ToNot(BeNil())
|
|
})
|
|
})
|