Podman 5.0.3 fails to start on M1 Mac

[erw444]

Issue Description

When starting Podman 5.0.3 on a M1 Mac, it just sits on the starting machine message.

Steps to reproduce the issue

Steps to reproduce the issue Note: My company has created a script to add our firewall certs to podman but below is our relevant configurations podman machine init podman machine stop podman machine set --cpus 8 --memory 5000 --disk-size 3000 podman machine set --rootful podman machine start

Describe the results you received

After running our configuration script we get the following: `Running podman machine init... Looking up Podman Machine image at quay.io/podman/machine-os:5.0 to create VM Getting image source signatures Copying blob 39d447e3fb35 done | Copying config 44136fa355 done | Writing manifest to image destination 39d447e3fb35b0626309d5e3591926284a4f1c370c2af5767b01350e3bf49d66 Extracting compressed file: podman-machine-default-amd64.raw: done
Machine init complete To start your machine run:

podman machine start

---

Machine "podman-machine-default" stopped successfully Starting machine "podman-machine-default"`

We also have tried to run podman machine start --log-level=debug: DEBU[0000] waiting for ready notification
DEBU[0000] vfkit command-line: [/opt/podman/bin/vfkit --cpus 8 --memory 5000 --bootloader efi,variable-store=/Users/pl3h269/.local/share/containers/podman/machine/applehv/efi-bl-podman-machine-default,create --device virtio-blk,path=/Users/pl3h269/.local/share/containers/podman/machine/applehv/podman-machine-default-amd64.raw --device virtio-rng --device virtio-serial,logFilePath=/var/folders/gd/dsmkb3x931q7xyn1fs8n3_bdxnq6k9/T/podman/podman-machine-default.log --device virtio-vsock,port=1025,socketURL=/var/folders/gd/dsmkb3x931q7xyn1fs8n3_bdxnq6k9/T/podman/podman-machine-default.sock,listen --device virtio-net,unixSocketPath=/var/folders/gd/dsmkb3x931q7xyn1fs8n3_bdxnq6k9/T/podman/podman-machine-default-gvproxy.sock,mac=5a:94:ef:e4:0c:ee --device virtio-fs,sharedDir=/Users,mountTag=a2a0ee2c717462feb1de2f5afd59de5fd2d8 --device virtio-fs,sharedDir=/private,mountTag=71708eb255bc230cd7c91dd26f7667a7b938 --device virtio-fs,sharedDir=/var/folders,mountTag=a0bb3a2c8b0b02ba5958b0576f0d6530e104 --restful-uri tcp://localhost:50089 --device virtio-gpu,width=800,height=600 --device virtio-input,pointing --device virtio-input,keyboard --gui --device virtio-vsock,port=1024,socketURL=/Users/pl3h269/.local/share/containers/podman/machine/applehv/podman-machine-default-ignition.sock,listen] DEBU[0000] ignition socket device: /Users/pl3h269/.local/share/containers/podman/machine/applehv/podman-machine-default-ignition.sock

A Fedora window opens when the log level is set and we get the following messages: Booting 'Fedora CoreOS 40.20240504.2.0 EFI stub: Decompressing Linux Kernel EFI stub: EFI_RNG_PROTOCOL unavailable

Describe the results you expected

Podman would start running. We have several other users running with M1 Macs running with Podman 5.0.3 working in our organization.

podman info output

OS: darwin/arm64
provider: applehv
version: 5.0.3

Cannot connect to Podman. Please verify your connection to the Linux system using `podman system connection list`, or try `podman machine init` and `podman machine start` to manage a new Linux VM
Error: unable to connect to Podman socket: failed to connect: dial tcp 127.0.0.1:49748: connect: connection refused

Podman in a container

No

Privileged Or Rootless

None

Upstream Latest Release

Yes

Additional environment details

Mac OS version: Sonoma 14.5 Macbook Pro 18 M1 10 cores 16GB of memory

Additional information

Additional information like issue happens only occasionally or issue happens with a particular architecture or on a particular setting

[ashley-cui]

Is this resolved with the latest podman? I'm unable to reproduce this.

[erw444]

Our enterprise has a months long process for pulling in new versions so we would like to get this working with 5.0.3.

[erw444]

Have pulled in 5.2.1 and it still is having this problem.

[jbluntz]

@ashley-cui any advice as to other diagnostics we could gather? As @erw444 mentioned only some users are affected by this issue and we've been trying to narrow down what might set those users apart, but so far haven't found anything conclusive. Are there other logs that we can provide that might point in the direction of the problem?

[jpedro1992]

I have the same problem:

A Fedora window opens when the log level is set and we get the following messages: Booting 'Fedora CoreOS 40.20240504.2.0 EFI stub: Decompressing Linux Kernel EFI stub: EFI_RNG_PROTOCOL unavailable

Any solution for this?

[stevelj]

@ashley-cui I work w/ @erw444 and @jbluntz. @jpedro1992's signature is exactly what we are seeing. When we put podman into debug mode it just seems the EFI Stub is not going anywhere. Is there any commands we could run for further debug? We've upgraded the few users we have with the issue to podman 5.2.1 and the issue persists. The linux command in the bootloader just hangs. We've been experiencing this with Intel MACs, Mx (1,2,3) as well. Can not seem to isolate a root cause.