colima: Colima won't start - log says 'error calling fd_connect'

Description

When I do colima start I get this:

> exiting, status={Running:false Degraded:false Exiting:true Errors:[] SSHLocalPort:0} (hint: see "/Users/joachim/.colima/_lima/colima/ha.stderr.log")

The last item in the log is this:

{"level":"fatal","msg":"template: :1:21: executing \"\" at \u003cfd_connect \"/Users/joachim/.colima/_lima/_networks/user-v2/user-v2_qemu.sock\"\u003e: error calling fd_connect: fd_connect: dial unix /Users/joachim/.colima/_lima/_networks/user-v2/user-v2_qemu.sock: connect: connection refused","time":"2023-12-20T10:10:40Z"}

I’ve tried restarting my mac but the error still occurs

Version

colima version 0.6.1 git commit: fa1bc249f921287745f01fb02075c9ecbb491c31 limactl version 0.18.0 qemu-img version 8.1.3 Copyright © 2003-2023 Fabrice Bellard and the QEMU Project developers

Operating System

  • macOS Intel <= 13 (Ventura)
  • macOS Intel >= 14 (Sonoma)
  • Apple Silicon <= 13 (Ventura) 13.5.1
  • Apple Silicon >= 14 (Sonoma)
  • Linux

Output of colima status

FATA[0000] colima is not running

Reproduction Steps

  1. Do colima start

Expected behaviour

No response

Additional context

No response

About this issue

  • Original URL
  • State: open
  • Created 3 years ago
  • Reactions: 28
  • Comments: 42 (2 by maintainers)

Most upvoted comments

I was able to fix such an issue by running rm -rf ~/.colima/_lima/_networks/user-v2. but still it should not occur in the first place.

I had this issue today, the error (if is the same) lies in lima, which now have an improved network stack more info

So I ended up deleting all (brew remove qemu lima colima && rm -rf ~/.lima && rm -rf ~/.colima) and reinstalling it…

Possibly removing only lima and reinstalling it could work… also apparently to this could be only a configuration file in lima; but since I didn’t care about deleting everything I went for the easy fix

Other solutions like re-installing or restarting didnt help in my case but removing ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid file worked for me

I was able to fix such an issue by running rm -rf ~/.colima/_lima/_networks/user-v2. but still it should not occur in the first place.

I’ve tried this, and it does work for me.

it might also be enough to just remove the .pid file in that folder. but to be safe removing the whole folder might be better.

Thank you @AndreasA ! I had the same problem after upgrading to MacOS Ventura 13.6.4. Simply removing _~/.colima/_lima/networks/user-v2/usernet_user-v2.pid solved the problem. You saved my day!

I was able to fix such an issue by running rm -rf ~/.colima/_lima/_networks/user-v2. but still it should not occur in the first place.

This resolved my issue 👍. Can somebody explain what this does?

it might also be enough to just remove the .pid file in that folder. but to be safe removing the whole folder might be better.

removing just the PID file worked for me, as a workaround. Would still love for this to stop happening, of course 😁 rm ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid

Actually, though, the problem came back after the next restart

😭

Deleting the whole directory rm -rf ~/.colima/_lima/_networks/user-v2/ resolved the error, but blew away my all my virtual machines and volumes, causing me to lose days of work…

😭 😭

it might also be enough to just remove the .pid file in that folder. but to be safe removing the whole folder might be better.

Other solutions like re-installing or restarting didnt help in my case but removing ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid file worked for me

Helped me too, thanks 😃

Had the same issue, installed and uninstalled a few times, restarted a few times, still no use.

This is what worked in the end (not sure if all steps are required exactly like this).

brew uninstall --zap qemu colima lima docker
brew autoremove
rm -rf ~/.colima && rm -rf ~/.lima && rm -rf ~/Library/Caches/lima

<restart>

brew update && brew upgrade
brew install lima
brew install --HEAD colima
brew install docker
colima start --cpu 4 --memory 6 --disk 100 --arch aarch64 --vm-type=vz
sudo ln -sf /Users/$USER/.colima/docker.sock /var/run/docker.sock

A bit brute force, but it worked.

it might also be enough to just remove the .pid file in that folder. but to be safe removing the whole folder might be better.

removing just the PID file worked for me, as a workaround. Would still love for this to stop happening, of course 😁

rm ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid

Thanks @AndreasA!

rm -rf ~/.colima/_lima/_networks/user-v2

Also worked for me! thank you!

it might also be enough to just remove the .pid file in that folder. but to be safe removing the whole folder might be better.

Removing only the pid file fixed this issue for me. Thanks @AndreasA!

All of these solutions, especially those that require removing and reinstalling colima are only fixing the symptoms but not the actual underlying issue.

I would love to see this resolved, but so far there doesn’t seem to be any progress or even recognition of the issue?

Other solutions like re-installing or restarting didnt help in my case but removing ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid file worked for me

This works, thank you!

Other solutions like re-installing or restarting didnt help in my case but removing ~/.colima/_lima/_networks/user-v2/usernet_user-v2.pid file worked for me

This work for me

I was able to fix such an issue by running rm -rf ~/.colima/_lima/_networks/user-v2. but still it should not occur in the first place.

It worked for me too, thanks.

@AndreasA Thank you! It solved my problem!

@AndreasA fix worked for me too 👍

Could Colima act on shutdown to delete that .pid file?

This is also reproducible in Apple Silicon >= 14 (Sonoma) Workaround of AndreasA seem to temporarily solve the issue.