Page 1 of 1

Discarding a snapshot in VirtualBox v2.0.2 hogs CPU. Normal?

Posted: 22. Sep 2008, 22:51
by ant
I don't recall having this problem in the past. I noticed deleting a snapshot hogs my CPU:

Code: Select all

top - 13:39:20 up 1 day,  9:38,  3 users,  load average: 7.34, 4.12, 1.97
Tasks: 120 total,   3 running, 117 sleeping,   0 stopped,   0 zombie
Cpu(s):  4.0%us, 22.7%sy, 10.3%ni,  0.0%id, 61.7%wa,  1.3%hi,  0.0%si,  0.0%st
Mem:   1035332k total,  1020288k used,    15044k free,     4188k buffers
Swap:  2361512k total,      124k used,  2361388k free,   723248k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                       
21896 ant       20   0 15640 6532 3648 S 10.3  0.6   0:31.89 VBoxSVC                       
 5154 ant       32  12 63268  44m 7012 R  5.7  4.4 126:54.00 compiz.real                   
 4739 root      20   0  143m  74m 9232 S  4.3  7.4 114:26.97 Xorg                          
 4854 ant       28   8 38336  12m 8440 S  3.7  1.2  46:30.21 gkrellm                       
21131 ant       20   0 69032  53m 3484 R  3.3  5.3   1:00.19 ruby                          
29041 root      20   0     0    0    0 S  1.3  0.0   0:00.16 pdflush                       
 5202 ant       28   8 89784  32m  18m S  1.0  3.2  34:46.72 amule                         
  154 root      15  -5     0    0    0 S  0.7  0.0   0:10.38 kswapd0                       
21875 ant       20   0 89952  22m  13m S  0.7  2.2   0:06.42 VirtualBox                    
28989 ant       20   0  2392 1148  884 R  0.7  0.1   0:00.36 top               
...
Is this normal? My desktop, mouse cursor, SSH to this box, etc. became unresponsive (almost frozen) and very slow. I know my machine is old as shown in http://alpha.zimage.com/~ant/antfarm/ab ... puters.txt (Secondary/Backup Computer, but I never had this problem before VirtualBox v2. Maybe it's a Kernel upgrade (had 2.6.22-K7) in the past) and many daily upgrades (apt-get upgrade). I am using virtualbox-2.0_2.0.2-36488_Debian_lenny_i386.deb from virtualbox.org so I hope that is the correct one to use (no errors). No errors in dmesg and hdparm, with the image, look fine to me:

Code: Select all

# hdparm /dev/hda

/dev/hda:
 multcount     =  0 (off)
 IO_support    =  1 (32-bit)
 unmaskirq     =  1 (on)
 using_dma     =  1 (on)
 keepsettings  =  0 (off)
 readonly      =  0 (off)
 readahead     = 256 (on)
 geometry      = 16383/255/63, sectors = 156301488, start = 0
Thank you in advance. :)

Posted: 22. Sep 2008, 22:57
by Sasquatch
The CPU should be powerful enough. Nothing wrong with that.

Are you sure that you discarded the snapshot and not merged it? Merging takes more CPU as it adds the data to the VDI (or previous snapshot).

Posted: 22. Sep 2008, 23:10
by ant
Sasquatch wrote:The CPU should be powerful enough. Nothing wrong with that.

Are you sure that you discarded the snapshot and not merged it? Merging takes more CPU as it adds the data to the VDI (or previous snapshot).
I right clicked on an old snapshot and clicked on "Discard Snapshot". Oh and this was the first snapshot in the list.

Posted: 22. Sep 2008, 23:43
by Sasquatch
As I don't use snapshots, I can only assume things. It's possible that the high CPU usage came from the fact that when you removed the first snapshot, it had to recalculate the tree and data integrity of the VM. The tree should be no problem, but the data integrity is something different.