Beta appears to hang when host sleeps

S

SusanMackay

Bit poster
I was running the Beta of parallels when I was called away for some time. In that time, the host went to 'sleep'. When it awoke, Parallels appears to be frozen, The guest system (Windows XP iin my case) doesn't respond to the mouse or keyboard.

However it uses about 50% of my CPU (I have a Macbook Pro with 2GHz Core Duo and 2GB RAM) whie the surrounding window is displayed and this drops to about 5% when the Parallels window is minimised. When the window is restored, the CPU usage returns to about 50%. Also all of the Paralles menu options are greyed out except the 'Services' and 'Hide...' options.

To my mind, this is a bug - nothing should stop the guest from responding, and Parallels should be sending the guest the approrpiate BIOS-type signals when the host sleeps or shuts down etc.

Is there a way out of this? At the moment all I can do is Force Quit Parallels and loose my recent work.

Thanks

Susan
 
I'm having the same problem with build 1898 on a Mac Pro tower (2GHz, 2GB). I tried to suspend the hung VM and the "Please wait" dialog came up but made no progess. When I Force Quit, I can't relaunch Parallels (Finder gives "unexpected error" code -600).

I'm attaching a sample from the Activity Monitor.
 

Attachments

I'm running RC2 on my Mac Pro and it brings down the whole OS if my Mac goes to sleep. It wakes up but then everything except the mouse cursor is frozen (Parallels, apps, Finder... everything).
 
I have had this problem since the first released version -- I've reported the bug twice but not recieved a reply. Its a *big* problem.

Note that disabling all of the power management features in WinXP doesn't seem to help, and it definately seems to happen more frequently if a rosetta app is open (eg Word.)
 
Sleep/freeze here too

I have seen the same thing several times here too - once in a while (not every time) when I am working in XP and the Mac Pro sleeps and wakes, XP is frozen. Only fix seems to be force quit Parallels. XP will restart without incident, only to lull me into a false sense of safety until it locks up again.

This build of Parallels also seems to have some issues with the GUI in a PCB design application I use, but it's not a show stopper.

No KPs here since 1922.

Any ideas, Parallels? Thanks!

Mac Pro 2.66 / 3 GB / 10.4.8
Parallels 1922 / XP Pro
 
Add me to the list, although I only noticed the problem beginning with 10.4.8 and build 1922. It happens on both my Mac Pro and my MacBook Pro.
 
Battery drain in sleep

I had this problem a few times, in different variations with a MBP 17" 2GB. Usually the effect is that either the MBP does not go to sleep (white LED on the lid lock not "breathing" but steady) or goes to sleep but does not awake, or is near-locked when it awakes (UI veeeery slow until I manage to force quit Parallels).

I am also suspecting (I say - suspecting! not accusing! I might be wrong!) that Parallels might be involved in draining my battery during (apparent) sleep. I noticed a few times that after an apparent sleep ("breathing" LED) overnight when the machine woke up it was quite warm and had drained the battery quite a bit.

I wouldn't mind too much if that happen once in a while - a glitch in complex software.

However, two times during such a "high powered" sleep, apparently the machine drained the battery to death. When I turned the machine back on, it indicated "no battery installed". The battery could not be recovered in both occasions and had to be replaced. Both times Parallels was running before the sleep and did not wake correctly afterwards (the MBP woke from HD based deep sleep when connected to AC).

Again: I don't blame Parallels for that, but I'm trying to find a pattern in the events that now ruined my battery twice, so I'm looking for similar experiences. My current guess is that there is a bug in Apple's OS/Firmware such that under special circumstances the MBP cannot power off even when the battery is already below the minimum capacity allowed and continues drawing power until the battery physically dies.
Parallels does do things at lowest level possible in the CPU (the VT-X hypervisor stuff), so it might be that it is involved in the "special circumstances". But even if Parallels was doing something wrong, I think that the SMC should be able to do an emergency shutdown even if the Core Duo CPU and the OS are completely locked up, and if it doesn't, it would be an Apple problem.

But not many people know as much about the low level Mac stuff as you Parallels guys do, so that's why I'm asking here...:)
 
jmayzurk said:
Add me to the list, although I only noticed the problem beginning with 10.4.8 and build 1922. It happens on both my Mac Pro and my MacBook Pro.
Click to expand...

Same here. I can't leave the MacBook for more than 10 minutes, or else as soon as the display sleeps, the guest OS completely freezes.
 
You must log in or register to reply here.
Back
Top