[OpenIndiana-discuss] Yikes my new install is frozeup in Vbox

[Harry Putnam]

Jim Klimov <jimklimov at cos.ru> writes:

> On 2013-08-07 03:02, Harry Putnam wrote:
>> Sorry about the line noise.  I always forget how much can be fixed on
>> windows by a reboot.  It must be the most often rebooted OS on the
>> planet.
>
>
> Well, quite often killing some offending processes can also do the
> trick ;) In case of VirtualBox, there are 3 or 4 VBox*.exe seen in
> the Task Manager, and if your VMs are stopped and GUI is closed
> these other helper processes should die off in a minute or so (not
> instantly, though - they wait a bit for their "clients" to reconnect).
> I guess something locked up in your processes, and these might need
> to be killed off via Task Manager.
>
> I believe you don't run VMs as services? There are also some bad
> DCOM interactions which forbid running interactive (GUI and CLI)
> and service'd VirtualBox programs simultaneously.

I'm actually running it as a practice run for a while with superfluous
data before plopping out $2000 $2500 for a more powerful win box with
a lot more ram.  I have 8GB on this laptop I'm using with older i7
820.  Expecting to get a mobo supporting 32gb and just buck up and get
the ram right from the start.  Probably an i7 or one of the heavy
hitting AMD cpus.

When I finally do setup... I hope to run a zfs server, and a heavyish
adobe inclined graphics workstation. The zfs server in vm.


> In VirtualBox per-machine folders there are also Logs/ where you
> can review causes for the recent crashes. Mostly the text is quite
> cryptic, sometimes just pointing to "Guru meditation", but there
> might still be clues... perhaps, inability to reserve a contiguous
> large stretch of virtual memory or somesuch.

I did look at those... but I couldn't tell poop about what I was
looking at.  None of it seemed to be about the freeze up.

> Finally, FWIW, on a Windows server running a VirtualBox with a
> Solaris installation, we had many problems with 4.2.* latest at
> that time, which were resolved by reverting to 4.1.24(?) latest

[...]

Thanks for the heads up about that.  I didn't notice anthing
particularly compelling about 4.2 so wouldn't mind backing off.

Its very good to know about things like this before making a costly
move ... thanks.