SVN 3125 & Hyper-V 2012 R2

[sudburr]

– svn: 3125
– Kernel: 3.19.1
– iPXE: 1.0.0+ (acc27)

I am unable to “Quick Register” or “Full Register” a Hyper-V 2012 R2 Generation 1 VM with FOG SVN 3125.

Quick Register sticks at “Attempting to register host …” plus slowly creeping CR+LF

Full Register throws up this:

Unable to register host for the following reasons:

BUG: using smp_processor_id() in preemptible [00000000] code: hdparm/2312
SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 00 00 00 00 20 00 00 … more 00
HDIO_GET_IDENTITY failed: Invalid argument

System Information: … lots of info …
BIOS Information: … lots of info …
Motherboard Information: … lots of info …

Then the screen clears and fills with a column of “None” and continues with Memory Info, then [B]Hard Disk: with no information attached[/B], then Case information.

Then it too sticks at “Attempting to register host …” plus slowly creeping CR+LF

The VM passes the Compatibility check.

I have verified that Quick and Full work on physical machines.

Concurrently, not seeing the HD prevents uploading even if I manually create the host in FOG.

I’m going to try 3127 now.

[sudburr]

– svn: 3127
– Kernel: 3.19.2
– iPXE: 1.0.0+ (acc27)

Problem still exists.

[Wayne Workman]

[quote=“sudburr, post: 44191, member: 4706”]-- svn: 3127
– Kernel: 3.19.2
– iPXE: 1.0.0+ (acc27)

Problem still exists.[/quote]

Try changing out the network adapter for the VM to something else.

I know in Win Server 12, you can choose something called “Legacy”.

Let us know how it goes.

[sudburr]

The VM is set to use a Legacy Adapter; or it wouldn’t have PXE booted at all as a Gen1 machine.

[sudburr]

I rolled the kernel back through successive older versions.

– svn: 3127
[B]-- Kernel: 3.18.5[/B]
– iPXE: 1.0.0+ (acc27)

… works.

[sudburr]

According to my FOG Configuration > Kernel Update page, the kernel has not been updated past 3.19.2.

Is a fix for this Hyper-V 2012 R2 problem in the queue of the list of many things to be done?

[Tom Elliott]

I guess I don’t know what the problem is?

[sudburr]

Look up, look way up… and I’ll call Rusty.

[Wayne Workman]

Oh. He’s referring to the original problem.

He’s looking for a fix so he doesn’t have to keep using that older kernel.

[url]https://books.google.com/books?id=8ecUy98-WqYC&pg=PA22&lpg=PA22&dq=Look+up,+look+way+up…+and+I%27ll+call+Rusty.&source=bl&ots=kGRsj1wxUm&sig=os_6reV1bFJGNxURDuzYLypzBVY&hl=en&sa=X&ei=Hm8dVfXvCZbhoASVooD4AQ&ved=0CEUQ6AEwBg#v=onepage&q=Look%20up%2C%20look%20way%20up…%20and%20I’ll%20call%20Rusty.&f=false[/url]

[sudburr]

Yes, the Friendly Giant. Good times …

[Sebastian Roth]

Sure you haven’t changed IDE controller in hyper V somewhere along the way??
[url]http://superuser.com/questions/874084/how-identify-a-virtual-hard-disk-how-unique-in-linux[/url]

Possibly a major change in the kernel is causing this. But then, wouldn’t heaps of other people have issues with this then too?? Sorry, don’t have hyper V, so I cannot test this on my own. Just rumbling…

[Wayne Workman]

I haven’t updated to the latest revision of FOG in about a month.

Generally, I only update when Tom asks me to, LOL! Although I think that might change… I’m building FOG on Fedora 21 server at home, and I will stay bleeding edge with that for a while.

[sudburr]

The problem still exists for [B]Kernel: 3.19.3 .[/B]

[Tom Elliott]

What, exactly, is the problem?

Are you getting the same
[code]BUG: using smp_processor_id() in preemptible [00000000] code: hdparm/2312[/code]

Or is there another message?

Can you enable the FOG_KERNEL_DEBUG and increase the Console level?

[sudburr]

A quick test of Kernel 4.0.0 presents a new failure. BTW I’m still using SVN 3127.

/bzImage… ok
/init.xz… ok
tulip: Cannot enable tuilip board #0, aborting
piix4_smbus 0000:00:07.3: SMBus base address uninitialized - upgrade BIOS or use force_addr=0xaddr
Starting logging: OK
Populating /dev using udev: done
Initializing random number generator… done.
Starting network…
ip: RTNETLINK answers: File exists
ssh-keygen: generating new host keys: RSA1 RSA DSA ECDSA ED25519
Starting sshd: OK

< screen clears >
< Free Computer Imaging Solution banner page appears followed by >

An error has been detected!

Failed to initialize disk

Computer will reboot in 1 minute.

Actually… 4.0.0 presents this error on other physical machines as well.

Forgive my foggy response… burning off a migraine here, but how do I run debug for quick/full host registration?

[Tom Elliott]

Update to latest svn then tell me of issues.

[sudburr]

Updating now… brb. wheeee

[sudburr]

Okay, looks like I pulled down 3283 and I have a new option upon installation:

What MySQL connection type would you like?
MYSQLI_STORE_RESULT is normal
MYSQL_ASYNC is what is Defaulted and recommended
Any value not equal to MYSQLI_ASYNC and not blank will use store result

To answer, what did FOG set up prior to this?

[Tom Elliott]

It’s been a hoopla of trials. I recommend leaving it blank/MYSQL_ASYNC

[sudburr]

Continuing the install with MYSQLI_ASYNC …

– svn: 3283
[B]-- Kernel: 4.0.0[/B]
– iPXE: 1.0.0+ (eda1)

The problem has reverted to as described at the top of this thread for the Hyper-V 2012 R2 client.

Reverting to [B]Kernel: 3.18.5 .[/B]