Applehv podman machine e2e "Basic ops" test timeout

[cevich]

Issue Description

When executing the basic podman machine e2e, every test hits 90s timeout.
Example (log).

Steps to reproduce the issue

Steps to reproduce the issue

1. brew tap cfergeau/crc
2. brew install go go-md2man coreutils vfkit cirruslabs/cli/cirrus
3. sudo sysadminctl -addUser test-user
4. sudo rm -rf /Users/test-user/*
5. sudo rm -rf /Users/test-user/.??*
6. sudo su - test-user
7. clone podman repo & cd into its directory
8. make .install.ginkgo podman-remote podman-mac-helper
9. export MACHINE_IMAGE="https://fedorapeople.org/groups/podman/testing/applehv/arm64/fedora-coreos-38.20230925.dev.0-applehv.aarch64.raw.gz"
10. export CONTAINERS_MACHINE_PROVIDER="applehv"
11. export GINKGO_TAGS="remote exclude_graphdriver_btrfs btrfs_noversion exclude_graphdriver_devicemapper containers_image_openpgp remote"
12. ./test/tools/build/ginkgo -vv --tags "$GINKGO_TAGS" -timeout=90m --trace --no-color --focus-file "basic_test.go" pkg/machine/e2e/.

Describe the results you received

run basic podman commands
/Users/test-user/podman/pkg/machine/e2e/basic_test.go:15
  Basic ops
  /Users/test-user/podman/pkg/machine/e2e/basic_test.go:28
  > Enter [BeforeEach] run basic podman commands - /Users/test-user/podman/pkg/machine/e2e/basic_test.go:21 @ 10/30/23 18:01:12.77
  < Exit [BeforeEach] run basic podman commands - /Users/test-user/podman/pkg/machine/e2e/basic_test.go:21 @ 10/30/23 18:02:34.581 (1m21.812s)
  > Enter [It] Basic ops - /Users/test-user/podman/pkg/machine/e2e/basic_test.go:28 @ 10/30/23 18:02:34.581
  /Users/test-user/podman/bin/darwin/podman machine init --image-path /private/tmp/fedora-coreos-38.20230925.dev.0-applehv.aarch64.raw.gz --now b829d29d93b8
  [FAILED] Timed out after 90.000s.
  Expected process to exit.  It did not.
  In [It] at: /Users/test-user/podman/pkg/machine/e2e/config_test.go:54 @ 10/30/23 18:04:04.635

Describe the results you expected

Basic tests should all pass

podman info output

OS: darwin/arm64
provider: qemu
version: 4.8.0-dev

Podman in a container

No

Privileged Or Rootless

Rootless

Upstream Latest Release

Yes

Additional environment details

Same results on MacOS 13 and 14

Additional information

Manually running ./bin/darwin/podman --log-level=debug machine init --image-path $MACHINE_IMAGE --now works perfectly fine.

I tried using the SpecTimeout decorator but couldn't get it to work - likely I wrote the golang bad/wrong.

[baude]

please consider a patch like this:

diff --git a/pkg/machine/e2e/basic_test.go b/pkg/machine/e2e/basic_test.go
index a1cfb71ef5..5fb543aed2 100644
--- a/pkg/machine/e2e/basic_test.go
+++ b/pkg/machine/e2e/basic_test.go
@@ -30,7 +30,7 @@ var _ = Describe("run basic podman commands", func() {
                // so skip it on cirrus envs and where CIRRUS_CI isn't set.
                name := randomString()
                i := new(initMachine)
-               session, err := mb.setName(name).setCmd(i.withImagePath(mb.imagePath).withNow()).run()
+               session, err := mb.setName(name).setCmd(i.withImagePath(mb.imagePath).withNow().withDiskSize(5)).run()
                Expect(err).ToNot(HaveOccurred())
                Expect(session).To(Exit(0))
 
@@ -58,7 +58,7 @@ var _ = Describe("run basic podman commands", func() {
        It("Podman ops with port forwarding and gvproxy", func() {
                name := randomString()
                i := new(initMachine)
-               session, err := mb.setName(name).setCmd(i.withImagePath(mb.imagePath).withNow()).run()
+               session, err := mb.setName(name).setCmd(i.withImagePath(mb.imagePath).withNow().withDiskSize(5)).run()
                Expect(err).ToNot(HaveOccurred())
                Expect(session).To(Exit(0))

[cevich]

Thanks, I'll try right away.

[cevich]

Unfortunately no love from that patch, tests fail in the same way 😢

Note: The "smoke" test takes a full (nearly) 9-minutes to run.. So if the basic test is doing a machine init, 90-seconds is nowhere near long enough.

Is there any way to get more info. out of the test? Maybe it's not failing in a way you/I assume? i.e. an actual hang somewhere vs something disk related.

[ashley-cui]

Closed via #20696