SATA error when VirtualBox is running

Discussions related to using VirtualBox on Linux hosts.
Post Reply
orest.kozyar
Posts: 2
Joined: 3. Sep 2008, 19:48

SATA error when VirtualBox is running

Post by orest.kozyar »

When I am running VirtualBox on Ubuntu Hardy Heron (8.04), the VM and other running applications (e.g. firefox, rhythmbox, etc) often hangs frequently for a few seconds, especially under conditions of heavy disk IO. When the system hangs, I see the following error messages repeating over and over:

[ 3999.283710] ata1.00: status: { DRDY ERR }
[ 3999.283712] ata1.00: error: { UNC }
[ 3999.287278] ata1.00: configured for UDMA/100
[ 3999.287289] ata1: EH complete
[ 4002.190218] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 4002.190225] ata1.00: irq_stat 0x40000001
[ 4002.190250] ata1.00: cmd c8/00:b8:c8:74:a0/00:00:00:00:00/e6 tag 0 dma 94208 in
[ 4005.096762] ata1.00: status: { DRDY ERR }
[ 4005.096765] ata1.00: error: { UNC }
[ 4005.100329] ata1.00: configured for UDMA/100
[ 4005.100357] Descriptor sense data with sense descriptors (in hex):
[ 4005.100413] ata1: EH complete

The only time I ever see these error messages is when VirtualBox is running. Any advice on how to fix this issue?
TerryE
Volunteer
Posts: 3572
Joined: 28. May 2008, 08:40
Primary OS: Ubuntu other
VBox Version: VirtualBox+Oracle ExtPack
Guest OSses: Ubuntu 10.04 & 11.10, both Svr&Wstn, Debian, CentOS
Contact:

Post by TerryE »

What version are you running? If you read the 1.6.6 announcement you will see a couple of the fixes are in this area.
Read the Forum Posting Guide
Google your Q site:VirtualBox.org or search for the answer before posting.
orest.kozyar
Posts: 2
Joined: 3. Sep 2008, 19:48

Post by orest.kozyar »

I'm running version 1.6.6.
aeichner
Oracle Corporation
Posts: 196
Joined: 31. Aug 2007, 19:12

Post by aeichner »

Just to be sure you get these errors on the host right? The fixes are for the emulated SATA controller in VirtualBox and wouldn't help there. I doubt that VirtualBox is the cause but probably triggers a bug in the kernel driver of your storage controller when having a high I/O load.
Can you try to reproduce this on another computer under the same conditions?
Post Reply