ubuntu 7.04 in Beta 4

[chuwy]

Has anybody used ubuntu 7.04 in the new beta 4? I have a mini project going on in a ubuntu vm and before i upgrade I would love to know others experiences, what features of the new beta apply to Ubuntu (linux) vm's? I assume that unity is windows only, although i would be delighted if it was part of the Linux tools :smileygrin:

[jared_oberhaus]

I just configured my Ubuntu 7.04 VM to run with vmi.present="TRUE". Here's what I found:

\* I got the dmesg output confirming Paravirtualized kernel.

\* Everything seems to have been working great

\* I didn't do any speed tests, but it certainly felt like it was fast.

\* I put my MacBook Pro to sleep, while the VM was running--note that I did not suspend my Ubuntu VM.

\* Woke up my MacBook Pro

\* Now my Ubuntu VM is brokenit appears that anything that does disk access completely freezes and doesn't do anything. So my Ubuntu VM is still running, but I can't open windows, save files, do anything. For instance, my emacs buffer was completely fine and responsive, until I tried to save the buffernow emacs is frozen. My xterm was working great until I tried to execute a command that read from the disk. And X was still working until I tried to open another window.

My guess is that there's some sort of entanglement in the vmi stuff that happens when the Mac is suspended. Just a guess...

Is there anything I should collect? Should I file a bug?

[admin]

I don't think there's anything to collect, but you should file a bug.

[skapplin]

I've confirmed your issue on my system just this minute. You should definitely file a bug.

[jared_oberhaus]

I just filed a support request, bug 189786381.

[skapplin]

I have filed one as well. I also tested without the vmi.present = "TRUE" in the .vmx file and the VM still freezes after waking from sleep. So this is a more general problem I think.

[schmidp]

Ubuntu 7.04 should just work.

Just for fun, consider adding vmi.present=TRUE to

your .vmx file. While we haven't tested it whatsoever

in VMware Fusion (we have done it extensively in

VMware Workstation 6), it might enable the

paravirtualized side of Ubuntu 7.04, making it the

faster Linux distro in a VMware VM.

Hi,

why dont't you put such things in the settings gui?

you could mark it as experimental if it's not tested yet.

lg philipp

[jared_oberhaus]

I haven't seen any such problem without vmi.present. Maybe you have to restart the entire VMware Fusion app to re-read the setting? I've noticed that my .vmx file gets re-set back to what it was originally if I don't have Fusion completely exited... And of course the setting won't work unless you reboot the actual guest OS, but obviously you must have done that in this case, as the OS was completely frozen...

[incognegro]

Ubuntu 7.04 should just work.

Just for fun, consider adding vmi.present=TRUE to

your .vmx file. While we haven't tested it whatsoever

in VMware Fusion (we have done it extensively in

VMware Workstation 6), it might enable the

paravirtualized side of Ubuntu 7.04, making it the

faster Linux distro in a VMware VM.

I got really excited about this. And I found out that this is only good for x86_32 and isn't supported for x86_64. And I happen to have installed x86_64. So here's hoping it saves someone some time trying to figure out why paravirtualization isn't working and why there are no Booting messages from dmesg mentioning paravirtualization.

\-- G.

[nospamboz]

enable the paravirtualized side of Ubuntu 7.04, making it the

faster Linux distro in a VMware VM.

I got really excited about this. And I found out that

this is only good for x86_32 and isn't supported for

x86_64.

Ubuntu 7.04 is based on Linux kernel 2.6.20, which only

supports VMI on i386 architecture. Ubuntu "Gutsy Gibbon"

in October will have a kernel that supports VMI in x86_64

as well, plus other good stuff. (I'm looking forward to CNR.)

[PhLy]

For the mouse problem, I can't see where is this work around by etung.

Can you be more explicit ?

[admin]

http://www.vmware.com/community/thread.jspa?messageID=614199&#614199

[PhLy]

Thanks you very much. I tried with little success.

I habe 2 VMs, both VM run X.org 7.2

-1- VM1 is Slax (=Slackware-current), no VMtools installed, but vmware_drv.so and vmmouse_drv.so provided by Xorg package.

-1.2 Without the workaround:

-a) works perfectly

-b) Xorg.0.log does report the problem :

(II) No default mouse found, adding one

(II) VMWARE(0): vmmouse is available

(**) Option "CorePointer"

(**) Mouse0: Core Pointer

(**) Option "Device" "/dev/input/mice"

(**) Option "ZAxisMapping" "4 5 6 7"

(**) Mouse0: ZAxisMapping: buttons 4 and 5

(WW) " (type: MOUSE)

(II) XINPUT: Adding extended input device "Mouse0" (type: MOUSE)

-1.2- With the workaround: mouse is very slow and not integrated with MacOS. So I gave up.

-2- VM2 is Ubuntu-7.10-T1 , paravirtualization enabled, no wmmouse_drv.so is available !

-2.1 Without VM tools: display OK but mouse as 1.2 above.

-2.2 With VM tools:

a) mouse is slow but it is integrated with MacOS

b)X.0.log doesn't report any mouse problem.

as network doesn't work any more I'll gave up VMtools.

[PhLy]

Enabled paravirtualization on Ubuntu-7.10T1.

I run this VM and another VM (Slax) without paravirtualization.

There is a huge difference on CPU consuming.

The paravirtualized VM consumes 5% when idle and the non paravirtualized one consumes 40%.

Can't say much more.

Ph.L.

[level323]

I can also confirm that I'm experiencing the same problems (with my Ubuntu guest OS) after my Mac host wakes from sleep.

I too am running an Ubuntu 7.04 VM (with paravirt enabled) on a 2GHz Macbook w/2GB RAM.

I can confirm that the problem DOES NOT occur if I explicitly suspend the Ubuntu guest (via the Fusion 'Virtual Machine' menu) before I send my Mac host to sleep. I can then 'wake' the Mac host and Resume the Ubuntu guest and the guest seems to come back up fine.

But the problem does DEFINITELY occur (and is repeatable) once the host wakes from sleep that was initiated while the Ubuntu guest was still running.

I'd call this a showstopper, not to mention a real pain. Looking forward to a solution.

I'm new to the VMTN forum - would anyone be so kind as to point me to the URL of the VMWare bug database where I can keep track of the bug # 189786381 that was previously raised concerning this problem?

Cheers

John

[Tony_Arnold]

Just as an update to my earlier experiences, I found that you do need to close Fusion after making the change - even if you shut down and start up your VM again. My CPU usage while booting went from ~40% to 0.5%. Very nice.