x/build: run DragonflyBSD VMs on GCE?

[bradfitz]

Looks like Dragonfly now supports virtio:

https://leaf.dragonflybsd.org/cgi/web-man?command=virtio&section=4

So it should run on GCE?

If somebody could prepare make.bash scripts to script the install to prepare bootable images, we could run it on GCE.

See the netbsd, openbsd, and freebsd directories as examples: https://github.com/golang/build/tree/master/env

(The script must run on Linux and use qemu to do the image creation.)

/cc @tdfbsd

[bradfitz]

More background on the dfly users list:
http://lists.dragonflybsd.org/pipermail/users/2017-December/313731.html

[bradfitz]

In that thread, @rickard-von-essen says:

I have a working packer build of DragonFly BSD https://github.com/boxcutter/bsd.

The most interesting parts are the boot_command
https://github.com/boxcutter/bsd/blob/master/dragonflybsd.json#L5
and actual installer script https://github.com/boxcutter/bsd/blob/master/http/install.sh.dfly

[bradfitz]

/cc @dmitshur

[bradfitz]

Update: I just ran Dragonfly (5.2.2) at home on QEMU/KVM with virtio-scsi and virtio net and it works fine.

So it should work fine on GCE, of course (which we already heard).

At this point I'm thinking we should just do this builder "by hand" for now, with a readme file with notes. I'll prepare the image by hand, then shut it down and copy its disk to a GCE image. (uploading it as a sparse tarball)

We can automate it with expect or whatnot later. Perfect is the enemy of good, etc.

[bradfitz]

I shut down my KVM/QEMU instance, copied its disk to a new GCE image, and created a GCE VM. It kernel panics on boot (over serial) with:

panic() at panic+0x236 0xffffffff805f8666 
panic() at panic+0x236 0xffffffff805f8666 
vfs_mountroot() at vfs_mountroot+0xfe 0xffffffff80672c7e 
mi_startup() at mi_startup+0x84 0xffffffff805c2a64 
Debugger("panic")
CPU0 stopping CPUs: 0x0000000e
 stopped
Stopped at      Debugger+0x7c:  movb    $0,0xe67a49(%rip)
db> 

So, uh, not as easy as I'd hoped.

[bradfitz]

Perhaps if we already have to do the whole double virtualization thing for Solaris (#15581 (comment)) anyway, we could just reuse that mechanism to run Dragonfly in qemu/kvm under GCE.

[cnst]

I've tried working on this earlier this year (back in 2018-02), and had it scripted to make the image automatically, but I had the same issue that it'd work on my machines with vanilla QEMU just fine, including with the disk being accessible on DFly through DragonFly's vtscsi(4) with a local QEMU as per the QEMU configuration magic described over at http://wiki.netbsd.org/tutorials/how_to_setup_virtio_scsi_with_qemu/, but it still wouldn't work on GCE with GCE's virtio_scsi. Is there any info on how GCE's virtio_scsi different from QEMU's virtio_scsi?

I've also tried running DragonFly BSD side by side with FreeBSD with CAMDEBUG, but it didn't seem to reveal anything obvious, although the underlying CAM logic does seem to be quite different, so, it's probably the one to blame. I didn't run out of ideas, but did ran out of time back in February, and recently my GCE credits ran out as well.

Nested virtualisation sounds interesting. Does it require Linux on GCE, or would FreeBSD also work?

[tuxillo]

@cnst do you have instructions on how you tried DragonFly on GCE?

[gopherbot]

Change https://golang.org/cl/162959 mentions this issue: dashboard, buildlet: add a disabled builder with nested virt, for testing

[tuxillo]

I've tried myself and it seems DragonFly is unable to find the disk.
We're working on it already: https://bugs.dragonflybsd.org/issues/3175

[gopherbot]

Change https://golang.org/cl/163057 mentions this issue: buildlet: change image name for COS-with-vmx buildlet

[gopherbot]

Change https://golang.org/cl/163301 mentions this issue: env/linux-x86-vmx: add new Debian host that's like Container-Optimized OS + vmx

[gopherbot]

Change https://golang.org/cl/202478 mentions this issue: dashboard: update Dragonfly tip policy for ABI change, add release builder

[bradfitz]

@tuxillo, looks like no progress on that bug, eh?

[tuxillo]

Thanks for the reminder, I kind of forgot about this one. It's being a tough one anyways. I'll check with the team again next week to see if we could do something.

[cnst]

@bradfitz I have some time to work on it again, but my credits expired, and trying to signup for a new account required some sort of an extra verification. Is there a way to get the credits again to work on this? Also, is there any way to reproduce this bug outside of Google environment? As per my 2018 comments, our driver works just fine in regular KVM using NetBSD's instructions for activating the codepath.

[bradfitz]

GCP has a Free Tier these days:
https://cloud.google.com/free/

COMPUTE

Compute Engine
1
F1-micro instance per month

Scalable, high-performance virtual machines.

1 f1-micro instance per month (US regions only — excluding Northern Virginia [us-east4])
30 GB-months HDD
5 GB-months snapshot in select regions
1 GB network egress from North America to all region destinations per month (excluding China and Australia)

There's no way to reproduce it locally. GCP uses KVM but doesn't use QEMU and its implementation of virtio-scsi etc isn't open source.

[cnst]

@bradfitz How long does it take recompile the kernel on this free instance? A few hours? It was already taking too long even on non-micro GCP instances compared to 15-year old hardware.

I think it'd be great if there was a way to reproduce this problem locally, because our virtio-scsi drivers work just fine with anything but the proprietary GCP implementation.

Would it be helpful to provide automation for any other cloud provider?

[bradfitz]

@cnst, I didn't imagine you'd be using the f1-micro installation for compilations. I thought you'd use your normal development environment to build and then use the f1-micro to test boot them on GCE until it worked.

[tuxillo]

@cnst what I did in my tests was to download the latest IMG, mount null it, build kernel with modifications and install it in the mountpoint. Then I used gcloud/gsutil to upload the img and create the disk and the instance. You can retrieve the console output with gcloud iirc.

[bradfitz]

because our virtio-scsi drivers work just fine with anything but the proprietary GCP implementation.

FWIW, Go runs the following operating systems on GCP that all work with Google's virtio-scsi implementation:

• Windows
• Linux
• OpenBSD
• FreeBSD
• NetBSD
• Plan 9

Either Dragonfly has a bug, or all those operating systems have worked around bugs in Google's implementation. Or both.

[tuxillo]

Just to give a quick update, we've done some steps in the right direction to fix this. At least the VM now sees the disk but further changes and testing are needed. I'll update this with more information as soonn as we have it.

da0 at vtscsi0 bus 0 target 1 lun 0
da0: <Google PersistentDisk 1> Fixed Direct Access SCSI-6 device 
da0: Serial Number                     
da0: 300.000MB/s transfers
da0: Command Queueing Enabled
da0: 2048MB (4194304 512 byte sectors: 255H 63S/T 261C)

[bradfitz]

Great!