Ticket #13764 (closed defect: obsolete)

Opened 5 years ago

Last modified 4 years ago

problems with multi physical processors server

Reported by: wassy83 Owned by:
Component: other Version: VirtualBox 4.3.20
Keywords: debian multi physical processors Cc:
Guest type: Windows Host type: Linux


System host DEBIAN 7 on IBM SERVER SYSTEM X with 2 physical CPU on board guest WINDOWS SERVER 2008 x64 r2 updated virtualbox running headless V 4.3.2 all extensions and addiction updated VTx enabled on 2x Intel(R) Xeon(R) CPU5160 @ 3.00GHz, dual core processor 3GB ram assigned 2x core assigned VDI is dinamically 250GB assigned

Problem: every time I'm trying to install an heavy software about to 6.8GB after 10 minutes system will abort, no bsod only black screen and machine istantly aborted, in general system seems to be slow and unstable even with simple desktop operations.

Fix: assigning one core fixed the instability: in a different office that are using same softwares, I replied the same identical configuration with no problem, only difference is the fauty one has 2x Intel(R) Xeon(R) CPU 5160 @ 3.00GHz, 2 cores the good one has 2x Intel(R) Xeon(R) CPU E5420 @ 2.50GHz, 4 cores

on the good one I have 2 guest "asterisk 2X core,server 2008 2X core" on the faulty one only guest is server 2008 that was running unstable with 2X core. I think that this is related with multi physical processors server Log is always the same and is attached

many thanks


wassy_log.txt Download (103.0 KB) - added by wassy83 5 years ago.
log after aborted machine

Change History

Changed 5 years ago by wassy83

log after aborted machine

comment:1 follow-up: ↓ 2 Changed 5 years ago by mhanor

What kind of storage device have you mounted at /mnt/dati ?
Anyway, I think that your crash is related to #13105.

comment:2 in reply to: ↑ 1 Changed 5 years ago by wassy83

Replying to mhanor:

What kind of storage device have you mounted at /mnt/dati ?
Anyway, I think that your crash is related to #13105.

hi and thank you for the reply, is a raid1 sata 500GB HD directly attached on my raid hw controller

comment:3 Changed 5 years ago by mhanor

Attach the VDI disk to a SAS or SCSI controller, to avoid the crash. Attach a log of that session, I think it would be interesting to know if those IO delays are still being logged. Are the RAID HDDs healthy? Are there any messages related to the RAID HDDs on the host OS logs, when VirtualBox crashes?

comment:4 Changed 5 years ago by wassy83

The raid hdd are new with no bad sectors, no related messages on the GUEST OS just black screen during heavy installations and guest OS aborted.. can't understand why are you talking about the drives, I fixed COMPLETELY the problem assigning one cpu to the guest OS so problem must be related to this, don't you think so? I have same configuration on another server with same RAID HW CONTROLLER and sata attached disks, only difference is that the good one is 2x quadcore and the faulty one is 2x dualcore. Yesterday I tried on the other server to assign 4 cores to one guest and system become extremely instable! in a multi phisical processor environment is like I have to assign a maximum of n-1 cores of a single processor to get a good stability.

comment:5 Changed 4 years ago by aeichner

  • Status changed from new to closed
  • Resolution set to obsolete

Please reopen if still relevant with a recent VirtualBox release.

Note: See TracTickets for help on using tickets.
ContactPrivacy policyTerms of Use