Problems with Hyper-V, NCQ SATA drives, system-start driver hvboot

[Guest Bruce Sanderson]

After installing Windows Server 2008 Enterprise 64 bit (RTM) and adding the

Hyper-V role, everything seemed to be fine - I could add and run Virtual

Machines.

 

I initially installed the RC 0 version of the HyperVisor -

Windows6.0-KB949219-x64.msu, then later the RC 1 update -

Windows6.0-KB950049-x64.msu.

 

I noticed there were many instances of System Event Log entries with Source

nvstor64, Event ID 129 with the text "Reset to device, \Device\RaidPort0,

was issued". I have not configured the disk controllers for RAID at all -

the RAID feature is disabled in the motherboard BIOS.

 

I noticed these after adding the Hyper-V role, but their frequency seemed to

increase significantly after installing the RC 1 update. This may be

becuase I used my Virtual Machines more, but this is not certain.

 

I have an ASUS P5N-D motherboard that has NVIDIA nforce 750i SLI chipset, 8

GB RAM and an Intel QuadCore Q6600 processor. The disks are:

Seagate ST332062 0AS (2 of)

Seagate ST332062 0AS (1 of) - the System/Boot partition is the first

partition on this drive

 

Device Manager in the "parent" VM reports these as "SCSI Disk Devices".

 

After a while I noticed that the system would every now and then "freeze"

for several seconds, even a minute or more, then continue as if nothing

happened. By "freeze" I mean"

- no mouse pointer movement when mouse moved

- Resource Monitor graphs etc. stop updating

- no response to any key strokes

When I checked the System Event log after the freeze ended, there would be

several "nvstor64 - 129" entries recorded, sometimes several inside of a

minute.

 

On the NVIDIA web site a found a reference to problems with NCQ hard drives

(http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=768&p_created=1138923863&p_sid=WrJKAz5j&p_accessibility=0&p_redirect=&p_lva=&p_sp=cF9zcmNoPSZwX3NvcnRfYnk9JnBfZ3JpZHNvcnQ9JnBfcm93X2NudD01NzcsNTc3JnBfcHJvZHM9JnBfY2F0cz0mcF9wdj0mcF9jdj0mcF9zZWFyY2hfdHlwZT1hbnN3ZXJzLnNlYXJjaF9ubCZwX3BhZ2U9MQ**&p_li=&p_topview=1,

or

http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_alp.php and

select nforce (motherboards) for Product, Troubleshooting for Category and

ncq for Keyword).

 

I can not find any "updated firmware" on the Seagate site for my disks. The

Seagate site says the "Serial ATA drives have NCQ enabled by default"

(http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_adp.php?p_faqid=2887&p_created=1149284511&p_sid=zdveCz5j&p_accessibility=0&p_redirect=&p_lva=&p_sp=cF9zcmNoPTEmcF9zb3J0X2J5PSZwX2dyaWRzb3J0PSZwX3Jvd19jbnQ9NSw1JnBfcHJvZHM9NDA5LDQxNSZwX2NhdHM9MCZwX3B2PTIuNDE1JnBfY3Y9JnBfc2VhcmNoX3R5cGU9YW5zd2Vycy5zZWFyY2hfbmwmcF9wYWdlPTEmcF9zZWFyY2hfdGV4dD1uY3E*&p_li=&p_topview=1

 

or

http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_alp.php?p_sid=v81xMz5j

and search for Product: Seagate, Serial ATA Drives, All Subs with the ncq in

the Search by Keword box).

 

I'm not an expert on these things, but I get the impression that the Native

Command Queuing (NCQ) feature is and has been part of the SATA specification

from the beginning and Seagate says "Since late 2004, most new SATA drive

families have supported NCQ", so it's a bit puzzling why a relatively new

chipset would have problems with this feature (the manual for the

motherboard is dated December, 2007 - I purchased it in May 2008). I

understand that Vista and Windows Server 2008 have built in driver support

for ncq.

 

Anyway, with these clue, in Device Manager, IDE ATA/ATAPI controller, NVIDIA

nForce Serial ATA Controller, Properties, Port 0 (and Port 1), I removed the

check mark from "Enable command queuing".

 

Now, I don't appear to be getting the "nvstor64 - 129" and the "freezing"

that seemed to be related.

 

However, every time Windows Server starts, I get this "Error" Event Log

entry:

Source: Service Control Manager Event Log Provider

Event ID 7026

General: "The following boot-start or system-start driver(s) failed to

load: hvboot"

 

Now, I have yet to discover anything that is NOT working - the HyperVisor

Manager works correctly, I can create, modify and run Virtual Machines

without any apparent problems.

 

So:

1. what is "hvboot"

2. what won't work when since it "failed to load"

3. what, if anything, can I do about this?

 

--

Bruce Sanderson

http://members.shaw.ca/bsanders

 

It is perfectly useless to know the right answer to the wrong question.

[Guest Meinolf Weber]

Re: Problems with Hyper-V, NCQ SATA drives, system-start driver hvboot

 

Hello Bruce,

 

Hvboot:

http://technet.microsoft.com/en-us/magazine/cc194386.aspx

 

Event id 7026:

http://technet2.microsoft.com/windowsserver2008/en/library/ee35505f-3522-414a-a778-a0da2f15e7251033.mspx?mfr=true

 

Best regards

 

Meinolf Weber

Disclaimer: This posting is provided "AS IS" with no warranties, and confers

no rights.

** Please do NOT email, only reply to Newsgroups

** HELP us help YOU!!! http://www.blakjak.demon.co.uk/mul_crss.htm

> After installing Windows Server 2008 Enterprise 64 bit (RTM) and

> adding the Hyper-V role, everything seemed to be fine - I could add

> and run Virtual Machines.

>

> I initially installed the RC 0 version of the HyperVisor -

> Windows6.0-KB949219-x64.msu, then later the RC 1 update -

> Windows6.0-KB950049-x64.msu.

>

> I noticed there were many instances of System Event Log entries with

> Source nvstor64, Event ID 129 with the text "Reset to device,

> \Device\RaidPort0, was issued". I have not configured the disk

> controllers for RAID at all - the RAID feature is disabled in the

> motherboard BIOS.

>

> I noticed these after adding the Hyper-V role, but their frequency

> seemed to increase significantly after installing the RC 1 update.

> This may be becuase I used my Virtual Machines more, but this is not

> certain.

>

> I have an ASUS P5N-D motherboard that has NVIDIA nforce 750i SLI

> chipset, 8

> GB RAM and an Intel QuadCore Q6600 processor. The disks are:

> Seagate ST332062 0AS (2 of)

> Seagate ST332062 0AS (1 of) - the System/Boot partition is the first

> partition on this drive

> Device Manager in the "parent" VM reports these as "SCSI Disk

> Devices".

>

> After a while I noticed that the system would every now and then

> "freeze"

> for several seconds, even a minute or more, then continue as if

> nothing

> happened. By "freeze" I mean"

> - no mouse pointer movement when mouse moved

> - Resource Monitor graphs etc. stop updating

> - no response to any key strokes

> When I checked the System Event log after the freeze ended, there

> would be

> several "nvstor64 - 129" entries recorded, sometimes several inside of

> a

> minute.

> On the NVIDIA web site a found a reference to problems with NCQ hard

> drives

> (http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php

> ?p_faqid=768&p_created=1138923863&p_sid=WrJKAz5j&p_accessibility=0&p_r

> edirect=&p_lva=&p_sp=cF9zcmNoPSZwX3NvcnRfYnk9JnBfZ3JpZHNvcnQ9JnBfcm93X

> 2NudD01NzcsNTc3JnBfcHJvZHM9JnBfY2F0cz0mcF9wdj0mcF9jdj0mcF9zZWFyY2hfdHl

> wZT1hbnN3ZXJzLnNlYXJjaF9ubCZwX3BhZ2U9MQ**&p_li=&p_topview=1,

> or

> http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_alp.php

> and

> select nforce (motherboards) for Product, Troubleshooting for Category

> and

> ncq for Keyword).

>

> I can not find any "updated firmware" on the Seagate site for my

> disks. The Seagate site says the "Serial ATA drives have NCQ enabled

> by default"

> (http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_adp.p

> hp?p_faqid=2887&p_created=1149284511&p_sid=zdveCz5j&p_accessibility=0&

> p_redirect=&p_lva=&p_sp=cF9zcmNoPTEmcF9zb3J0X2J5PSZwX2dyaWRzb3J0PSZwX3

> Jvd19jbnQ9NSw1JnBfcHJvZHM9NDA5LDQxNSZwX2NhdHM9MCZwX3B2PTIuNDE1JnBfY3Y9

> JnBfc2VhcmNoX3R5cGU9YW5zd2Vycy5zZWFyY2hfbmwmcF9wYWdlPTEmcF9zZWFyY2hfdG

> V4dD1uY3E*&p_li=&p_topview=1

>

> or

> http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_alp.ph

> p?p_sid=v81xMz5j and search for Product: Seagate, Serial ATA Drives,

> All Subs with the ncq in the Search by Keword box).

>

> I'm not an expert on these things, but I get the impression that the

> Native Command Queuing (NCQ) feature is and has been part of the SATA

> specification from the beginning and Seagate says "Since late 2004,

> most new SATA drive families have supported NCQ", so it's a bit

> puzzling why a relatively new chipset would have problems with this

> feature (the manual for the motherboard is dated December, 2007 - I

> purchased it in May 2008). I understand that Vista and Windows Server

> 2008 have built in driver support for ncq.

>

> Anyway, with these clue, in Device Manager, IDE ATA/ATAPI controller,

> NVIDIA nForce Serial ATA Controller, Properties, Port 0 (and Port 1),

> I removed the check mark from "Enable command queuing".

>

> Now, I don't appear to be getting the "nvstor64 - 129" and the

> "freezing" that seemed to be related.

>

> However, every time Windows Server starts, I get this "Error" Event

> Log

> entry:

> Source: Service Control Manager Event Log Provider

> Event ID 7026

> General: "The following boot-start or system-start driver(s) failed

> to

> load: hvboot"

> Now, I have yet to discover anything that is NOT working - the

> HyperVisor Manager works correctly, I can create, modify and run

> Virtual Machines without any apparent problems.

>

> So:

> 1. what is "hvboot"

> 2. what won't work when since it "failed to load"

> 3. what, if anything, can I do about this?

> It is perfectly useless to know the right answer to the wrong

> question.

>

[Guest Bruce Sanderson]

Re: Problems with Hyper-V, NCQ SATA drives, Reset to device, \Device\RaidPort0

 

Re: Problems with Hyper-V, NCQ SATA drives, Reset to device, \Device\RaidPort0

 

Thanks for the links, Meinolf.

 

That answers the first question but not the other two:

2. what won't work when since it "failed to load"

3. what, if anything, can I do about this?

 

I did a bit more investigating.

 

1. (since 6 May when the Hyper-V Role was added) the System Event Log has

52 occurrences of Source: EventLog, ID 6005 which are recorded every time

the system is started (or restarted). This is a test/experimental system

and is only running when it is in use, not continuously.

2. (since 6 May when the Hyper-V Role was added) the System Event Log has 52

occurrences of Source: Hyper-V-Hypervisor, ID 1 "Hyper-V successfully

started".

3. sometimes, the EventLog 6005 entry has a later time than the

corresponding Hype-V-Hypervisor 1 entry and sometimes it is earlier

(according to the time stamps on the Event Log entries)

4. There are only four occurrences of the Service Control Manager Event Log

Provider Event ID 7026 "The following boot-start or system-start driver(s)

failed to load: hvboot"

 

So whatever that problem is (hvboot not starting), it is very intermittent

and is probably not related to the disk I/O issue that is my main concern.

 

For the time being, in Device Manager, IDE ATA/ATAPI controller, NVIDIA

nForce Serial ATA Controller, Properties, Port 0 (and Port 1), I removed the

check mark from "Enable command queuing", which avoids the "freezing" and

the

"Reset to device" Event Log entries, but appears to significantly degrade

disk I/O performance when multiple Virtual Machines are running.

 

Is anyone aware of a way to cure the problem reported by the System Event

Log entries with Source nvstor64, Event ID 129 with the text "Reset to

device, \Device\RaidPort0, was issued", which appears to be the cause of the

system "freezing"?

 

 

--

Bruce Sanderson

http://members.shaw.ca/bsanders

 

It is perfectly useless to know the right answer to the wrong question.

 

 

 

"Meinolf Weber" <meiweb(nospam)@gmx.de> wrote in message

news:ff16fb669f7668ca95cb98a29c61@msnews.microsoft.com...

> Hello Bruce,

>

> Hvboot:

> http://technet.microsoft.com/en-us/magazine/cc194386.aspx

>

> Event id 7026:

> http://technet2.microsoft.com/windowsserver2008/en/library/ee35505f-3522-414a-a778-a0da2f15e7251033.mspx?mfr=true

>

> Best regards

>

> Meinolf Weber

> Disclaimer: This posting is provided "AS IS" with no warranties, and

> confers no rights.

> ** Please do NOT email, only reply to Newsgroups

> ** HELP us help YOU!!! http://www.blakjak.demon.co.uk/mul_crss.htm

>

>> After installing Windows Server 2008 Enterprise 64 bit (RTM) and

>> adding the Hyper-V role, everything seemed to be fine - I could add

>> and run Virtual Machines.

>>

>> I initially installed the RC 0 version of the HyperVisor -

>> Windows6.0-KB949219-x64.msu, then later the RC 1 update -

>> Windows6.0-KB950049-x64.msu.

>>

>> I noticed there were many instances of System Event Log entries with

>> Source nvstor64, Event ID 129 with the text "Reset to device,

>> \Device\RaidPort0, was issued". I have not configured the disk

>> controllers for RAID at all - the RAID feature is disabled in the

>> motherboard BIOS.

>>

>> I noticed these after adding the Hyper-V role, but their frequency

>> seemed to increase significantly after installing the RC 1 update.

>> This may be becuase I used my Virtual Machines more, but this is not

>> certain.

>>

>> I have an ASUS P5N-D motherboard that has NVIDIA nforce 750i SLI

>> chipset, 8

>> GB RAM and an Intel QuadCore Q6600 processor. The disks are:

>> Seagate ST332062 0AS (2 of)

>> Seagate ST332062 0AS (1 of) - the System/Boot partition is the first

>> partition on this drive

>> Device Manager in the "parent" VM reports these as "SCSI Disk

>> Devices".

>>

>> After a while I noticed that the system would every now and then

>> "freeze"

>> for several seconds, even a minute or more, then continue as if

>> nothing

>> happened. By "freeze" I mean"

>> - no mouse pointer movement when mouse moved

>> - Resource Monitor graphs etc. stop updating

>> - no response to any key strokes

>> When I checked the System Event log after the freeze ended, there

>> would be

>> several "nvstor64 - 129" entries recorded, sometimes several inside of

>> a

>> minute.

>> On the NVIDIA web site a found a reference to problems with NCQ hard

>> drives

>> (http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php

>> ?p_faqid=768&p_created=1138923863&p_sid=WrJKAz5j&p_accessibility=0&p_r

>> edirect=&p_lva=&p_sp=cF9zcmNoPSZwX3NvcnRfYnk9JnBfZ3JpZHNvcnQ9JnBfcm93X

>> 2NudD01NzcsNTc3JnBfcHJvZHM9JnBfY2F0cz0mcF9wdj0mcF9jdj0mcF9zZWFyY2hfdHl

>> wZT1hbnN3ZXJzLnNlYXJjaF9ubCZwX3BhZ2U9MQ**&p_li=&p_topview=1,

>> or

>> http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_alp.php

>> and

>> select nforce (motherboards) for Product, Troubleshooting for Category

>> and

>> ncq for Keyword).

>>

>> I can not find any "updated firmware" on the Seagate site for my

>> disks. The Seagate site says the "Serial ATA drives have NCQ enabled

>> by default"

>> (http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_adp.p

>> hp?p_faqid=2887&p_created=1149284511&p_sid=zdveCz5j&p_accessibility=0&

>> p_redirect=&p_lva=&p_sp=cF9zcmNoPTEmcF9zb3J0X2J5PSZwX2dyaWRzb3J0PSZwX3

>> Jvd19jbnQ9NSw1JnBfcHJvZHM9NDA5LDQxNSZwX2NhdHM9MCZwX3B2PTIuNDE1JnBfY3Y9

>> JnBfc2VhcmNoX3R5cGU9YW5zd2Vycy5zZWFyY2hfbmwmcF9wYWdlPTEmcF9zZWFyY2hfdG

>> V4dD1uY3E*&p_li=&p_topview=1

>>

>> or

>> http://seagate.custhelp.com/cgi-bin/seagate.cfg/php/enduser/std_alp.ph

>> p?p_sid=v81xMz5j and search for Product: Seagate, Serial ATA Drives,

>> All Subs with the ncq in the Search by Keword box).

>>

>> I'm not an expert on these things, but I get the impression that the

>> Native Command Queuing (NCQ) feature is and has been part of the SATA

>> specification from the beginning and Seagate says "Since late 2004,

>> most new SATA drive families have supported NCQ", so it's a bit

>> puzzling why a relatively new chipset would have problems with this

>> feature (the manual for the motherboard is dated December, 2007 - I

>> purchased it in May 2008). I understand that Vista and Windows Server

>> 2008 have built in driver support for ncq.

>>

>> Anyway, with these clue, in Device Manager, IDE ATA/ATAPI controller,

>> NVIDIA nForce Serial ATA Controller, Properties, Port 0 (and Port 1),

>> I removed the check mark from "Enable command queuing".

>>

>> Now, I don't appear to be getting the "nvstor64 - 129" and the

>> "freezing" that seemed to be related.

>>

>> However, every time Windows Server starts, I get this "Error" Event

>> Log

>> entry:

>> Source: Service Control Manager Event Log Provider

>> Event ID 7026

>> General: "The following boot-start or system-start driver(s) failed

>> to

>> load: hvboot"

>> Now, I have yet to discover anything that is NOT working - the

>> HyperVisor Manager works correctly, I can create, modify and run

>> Virtual Machines without any apparent problems.

>>

>> So:

>> 1. what is "hvboot"

>> 2. what won't work when since it "failed to load"

>> 3. what, if anything, can I do about this?

>> It is perfectly useless to know the right answer to the wrong

>> question.

>>

>

>