Event ID 9: did not respond within the timeout period

[ToddAndMargo]

host: centos 5.5 x64, 32 GB memory guest: windows server 2003, sp2, r2 32 bit, 4 GB memory VM: VB 3.2.12

Hi All,

Sorry for coming directly to you guys with this, but since I can not purchase support from you guys, I have no other choice. And, this is a real emergency. I have got to get this fixed or my goose is cooked (as in fired).

I am in trouble here. I just got a Fox Pro database corruption. The SATA controller was giving me Event: 9 iaStor error out the nose, so I switched our controller from an "IATA89ENU v8.9.0.1023" (SATA) controller to a "Intel 82371AB/EB 5.2.3790.3959" (IDE/ATAPI) controller. It stopped the iaStor errors. But now I get the same Event:9 ATAPI errors.

I HAVE TO STOP THESE ERRORS!

Log files will follow shortly.

Many thanks, -T

Event Type: Error Event Source: atapi Event Category: None Event ID: 9 Date: 3/2/2011 Time: 10:21:14 AM User: N/A Computer: VB-PPRO2 Description: The device, \Device\Ide\IdePort0, did not respond within the timeout period.

[ToddAndMargo]

logs for event 9

[ToddAndMargo]

I am finding this error all over my other Windows guests. Not a *flood* like the guest running M$SQL Server 2008, but here and there:

Event Type: Error Event Source: iaStor Event Category: None Event ID: 9 Date: 1/3/2011 Time: 1:54:23 PM User: N/A Computer: VB-WINXP Description: The device, \Device\Ide\iaStor0, did not respond within the timeout period.

Also: a work around to help you guys troubleshoot this issue:

Turn the Virtual Machine's hard drive cache off. Device Manager --> Disk drives --> VBOX HARDDRIVE --> Policies (tab) --> uncheck “Enable write caching on the disk”

This stopped the iaStor and ATAPI Event 9 errors. You guys now have something to fix.

-T

[Frank Mehnert]

Any chance you test ​this build? We fixed a guest corruption bug for SATA guests with high amount of guest RAM. There is a slight chance that your problem was related to that bug.

[ToddAndMargo]

Hi Frank,

This bug fix sounds exciting and like it may fix the problem. Unfortunately, I can not test this fix as I no longer have access to this equipment: the customer fired me. This bug royally corrupted both an M$SQL 2008 and a Fox Pro database. I never managed to regain my customer confidence after that, even though I came up with a work around (uncheck “Enable write caching on the disk” ). This bug shows up in droves under M$SQL 2008.

Thank you for the help anyway, -T

[Frank Mehnert]

Thanks, I so I will close this ticket.

[ToddAndMargo]

Hi Frank,

Just out of curiosity, on what do you base closing the ticket? Is it because I can not test the patch and you do not like having bugs hanging around? Or, did you reproduce the problem and the patch fixed it?

Many thanks, -T

[Frank Mehnert]

If you cannot test the fix then it does not make sense to keep it open. If you can reproduce the problem, feel free to reopen it.

[ToddAndMargo]

Hi Frank,

Okay, I see. Labeling this as "fixed" is incorrect. You need to label it with something that correctly describes the situation. Do you have a "won't fix" or "unable to duplicate" or some such? Anything other than "fixed", which it is clearly not.

-T

[Frank Mehnert]

I agree that fixed is not correct. OTOH, the current trac system does not provide an "unable to reproduce". But if someone has a similar issue he will read the last comments of this ticket and reopen it so I think this is not a big issue.

[Artem S. Tashkinov]

It's probably a Virtual Box 4.1.6 regression but this issue has resurfaced, I can reproduce it with Windows XP SP2,

iastor.sys 7.6.1.1002 iastor.inf 7.6.1.1002

It takes up to 2 minutes to initialize SATA HDD (real one, using a vmdk file) using this driver.

[Artem S. Tashkinov]

The entire virtual machine, excluding a bootable Windows XP SP2 BartPE CD-Rom

[Artem S. Tashkinov]

Actually my problem is different - it takes up to two (2!!) minutes to initialize a SATA HDD, but then everything more or less works.

Shutdown also takes ages to complete.

[Artem S. Tashkinov]

I've filed a new ticket 9859.

Please, close this one.