[utils] failed to run '[ -e /var/lib/cloud/instance/boot-finished ]

[marcblanchet]

Describe the bug
vm creation takes significant time. No apparent reason but logs show zillions of the following error message, printed every second. When the timeout is set, then those logs are stopped. So it looks like these logs are actually showing the culprit.

[debug] [utils] failed to run '[ -e /var/lib/cloud/instance/boot-finished ]'
[debug] [ssh session] Executing '[ -e /var/lib/cloud/instance/boot-finished ]'

To Reproduce

1. multipass ...

Expected behavior
should be created faster.

Logs
see above

Additional info

• OS: macOS Sequoia 15.2

• multipass version
  multipass 1.15.0+Mac
  multipassd 1.15.0+mac

• multipass info
  Name: GND
  State: Running
  Snapshots: 0
  IPv4: 192.168.66.94
  192.168.40.1
  Release: Ubuntu 24.10
  Image hash: fb39312ffd2b (Ubuntu 24.10)
  CPU(s): 1
  Load: 0.00 0.65 0.97
  Disk usage: 4.4GiB out of 19.3GiB
  Memory usage: 264.2MiB out of 951.5MiB
  Mounts: --

• multipass get local.driver
  qemu
  Additional context
  Add any other context about the problem here.

[ricab]

Hi @marcblanchet, those logs are normal debug-level messages showing the process of multipass polling the instance while waiting for cloud-init to finish setting it up. The launch command only completes once cloud-init finishes (or if the timeout is reached) but, at that point, the instance is already running and you should be able to shell into it (from a different terminal or after Ctrl+C). You can follow cloud-init logs inside the instance to see what it is doing.

Is there something specific where you think Multipass is at fault here?