Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Podman Machine VM should not auto update and reboot without user interaction #20122

Closed
tmoschou opened this issue Sep 25, 2023 · 3 comments · Fixed by #20190
Closed

Podman Machine VM should not auto update and reboot without user interaction #20122

tmoschou opened this issue Sep 25, 2023 · 3 comments · Fixed by #20190
Assignees
Labels
kind/bug Categorizes issue or PR as related to a bug. locked - please file new issue/PR Assist humans wanting to comment on an old issue or PR with locked comments. machine macos MacOS (OSX) related remote Problem is in podman-remote

Comments

@tmoschou
Copy link

Issue Description

We have made the switch from Docker Desktop for MacOS to Podman / Podman Desktop for nearly a year now. Unfortunately, on a too frequent basis, we experience issues with the stability of the Podman Machine VM (MacOS). Most I think can be attribute to breaking changes from pushed auto updates (via Zincati) in particular compatibility with docker / docker-compose.

Users are reporting they are having to either kill the VM if it gets into a bricked state (seems to be gvproxy socket unresponsive / can't ssh into machine) or destroy and recreate the VM, loosing any data/state.

Can I please suggest at a minimum Podman, via likely the ignition file, disables Zincati from auto updating the VM and rebooting while containers are in use and without prompt. https://coreos.github.io/zincati/usage/auto-updates/#disabling-auto-updates.

Examples of bugs impacted by Zincati auto updating and rebooting the local VM.

#19930 is particularly bad, I am advising users to rollback and/or use an older CoreOS ISO.

podman machine ssh
sudo rpm-ostree rollback
sudo systemctl disable --now zincati.service
sudo systemctl reboot
❯  podman version
Client:       Podman Engine
Version:      4.6.1
API Version:  4.6.1
Go Version:   go1.20.7
Git Commit:   f3069b3ff48e30373c33b3f5976f15abf8cfee20
Built:        Fri Aug 11 03:43:43 2023
OS/Arch:      darwin/arm64

Server:       Podman Engine
Version:      4.6.2
API Version:  4.6.2
Go Version:   go1.20.7
Built:        Wed Sep 13 05:37:26 2023
OS/Arch:      linux/arm64

Steps to reproduce the issue

Steps to reproduce the issue

  1. podman machine init
  2. run some containers and workloads
  3. wait for VM to reboot or get into a bad state

Describe the results you received

  • local VM reboots
  • gvproxy can be unresponsible
  • podman machine ssh doesn't work
  • containers are re-started
  • host file mounts are lost

Describe the results you expected

local VM is not autoupdated / rebooted without user interaction

podman info output

host:
  arch: arm64
  buildahVersion: 1.31.2
  cgroupControllers:
  - cpu
  - io
  - memory
  - pids
  cgroupManager: systemd
  cgroupVersion: v2
  conmon:
    package: conmon-2.1.7-2.fc38.aarch64
    path: /usr/bin/conmon
    version: 'conmon version 2.1.7, commit: '
  cpuUtilization:
    idlePercent: 99.98
    systemPercent: 0.01
    userPercent: 0.01
  cpus: 4
  databaseBackend: boltdb
  distribution:
    distribution: fedora
    variant: coreos
    version: "38"
  eventLogger: journald
  freeLocks: 2033
  hostname: localhost.localdomain
  idMappings:
    gidmap:
    - container_id: 0
      host_id: 1000
      size: 1
    - container_id: 1
      host_id: 100000
      size: 1000000
    uidmap:
    - container_id: 0
      host_id: 503
      size: 1
    - container_id: 1
      host_id: 100000
      size: 1000000
  kernel: 6.4.15-200.fc38.aarch64
  linkmode: dynamic
  logDriver: journald
  memFree: 7738359808
  memTotal: 8299999232
  networkBackend: netavark
  networkBackendInfo:
    backend: netavark
    dns:
      package: aardvark-dns-1.7.0-1.fc38.aarch64
      path: /usr/libexec/podman/aardvark-dns
      version: aardvark-dns 1.7.0
    package: netavark-1.7.0-1.fc38.aarch64
    path: /usr/libexec/podman/netavark
    version: netavark 1.7.0
  ociRuntime:
    name: crun
    package: crun-1.9-1.fc38.aarch64
    path: /usr/bin/crun
    version: |-
      crun version 1.9
      commit: a538ac4ea1ff319bcfe2bf81cb5c6f687e2dc9d3
      rundir: /run/user/503/crun
      spec: 1.0.0
      +SYSTEMD +SELINUX +APPARMOR +CAP +SECCOMP +EBPF +CRIU +LIBKRUN +WASM:wasmedge +YAJL
  os: linux
  pasta:
    executable: /usr/bin/pasta
    package: passt-0^20230908.g05627dc-1.fc38.aarch64
    version: |
      pasta 0^20230908.g05627dc-1.fc38.aarch64-pasta
      Copyright Red Hat
      GNU General Public License, version 2 or later
        <https://www.gnu.org/licenses/old-licenses/gpl-2.0.html>
      This is free software: you are free to change and redistribute it.
      There is NO WARRANTY, to the extent permitted by law.
  remoteSocket:
    exists: true
    path: /run/user/503/podman/podman.sock
  security:
    apparmorEnabled: false
    capabilities: CAP_CHOWN,CAP_DAC_OVERRIDE,CAP_FOWNER,CAP_FSETID,CAP_KILL,CAP_NET_BIND_SERVICE,CAP_SETFCAP,CAP_SETGID,CAP_SETPCAP,CAP_SETUID,CAP_SYS_CHROOT
    rootless: true
    seccompEnabled: true
    seccompProfilePath: /usr/share/containers/seccomp.json
    selinuxEnabled: true
  serviceIsRemote: true
  slirp4netns:
    executable: /usr/bin/slirp4netns
    package: slirp4netns-1.2.1-1.fc38.aarch64
    version: |-
      slirp4netns version 1.2.1
      commit: 09e31e92fa3d2a1d3ca261adaeb012c8d75a8194
      libslirp: 4.7.0
      SLIRP_CONFIG_VERSION_MAX: 4
      libseccomp: 2.5.3
  swapFree: 0
  swapTotal: 0
  uptime: 20h 7m 18.00s (Approximately 0.83 days)
plugins:
  authorization: null
  log:
  - k8s-file
  - none
  - passthrough
  - journald
  network:
  - bridge
  - macvlan
  - ipvlan
  volume:
  - local
registries:
  search:
  - docker.io
store:
  configFile: /var/home/core/.config/containers/storage.conf
  containerStore:
    number: 9
    paused: 0
    running: 0
    stopped: 9
  graphDriverName: overlay
  graphOptions: {}
  graphRoot: /var/home/core/.local/share/containers/storage
  graphRootAllocated: 53082042368
  graphRootUsed: 23652081664
  graphStatus:
    Backing Filesystem: xfs
    Native Overlay Diff: "true"
    Supports d_type: "true"
    Using metacopy: "false"
  imageCopyTmpDir: /var/tmp
  imageStore:
    number: 33
  runRoot: /run/user/503/containers
  transientStore: false
  volumePath: /var/home/core/.local/share/containers/storage/volumes
version:
  APIVersion: 4.6.2
  Built: 1694549246
  BuiltTime: Wed Sep 13 05:37:26 2023
  GitCommit: ""
  GoVersion: go1.20.7
  Os: linux
  OsArch: linux/arm64
  Version: 4.6.2

Podman in a container

No

Privileged Or Rootless

Rootless

Upstream Latest Release

Yes

Additional environment details

Additional environment details

Additional information

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

@tmoschou tmoschou added the kind/bug Categorizes issue or PR as related to a bug. label Sep 25, 2023
@github-actions github-actions bot added macos MacOS (OSX) related remote Problem is in podman-remote labels Sep 25, 2023
@afbjorklund
Copy link
Contributor

@Luap99 Luap99 added the machine label Sep 25, 2023
@baude baude self-assigned this Sep 28, 2023
@baude
Copy link
Member

baude commented Sep 28, 2023

@tmoschou thanks for the excellent write up ... very appreciated.

@cgwalters
Copy link
Contributor

Persistently disabling auto updates is documented at https://coreos.github.io/zincati/usage/auto-updates/#disabling-auto-updates

Alternatively, one can keep zincati.service disabled by default, but only systemctl start it when desired.

baude added a commit to baude/podman that referenced this issue Sep 28, 2023
As explained in containers#21022, there are all kinds of downsides to a machine
updating itself (via zincati) automatically, like interuption of
service, lost mounts, etc.

disabling zincati will at least allow stop these downsides.  we are
likely to contemplate if podman will take over the update process
externally where interuption of services will not occur etc.

Fixes containers#20122

Signed-off-by: Brent Baude <[email protected]>
@github-actions github-actions bot added the locked - please file new issue/PR Assist humans wanting to comment on an old issue or PR with locked comments. label Dec 29, 2023
@github-actions github-actions bot locked as resolved and limited conversation to collaborators Dec 29, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
kind/bug Categorizes issue or PR as related to a bug. locked - please file new issue/PR Assist humans wanting to comment on an old issue or PR with locked comments. machine macos MacOS (OSX) related remote Problem is in podman-remote
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants