Currently I have the need to build a new Windows XP Guest in VirtualBox. So I figure that I want the latest and greatest so that I wont have to keep upgrading it.
On the OS side, I want it to be pre-installed with SP3, and possibly some or all of the patches
On the Virtual Hardware side I want for the Hard drive to be SATA, not IDE, and for the bridge to be ICH9, not PXII3.
My ultimate goal is for it to be point and shoot, no installing as IDE / PXII3 and convert, but to choose SATA / ICH9 straight from the start. Also want to do it without any other tools like nlite. Starting work today, lets see where this gets us
Sunday, July 24, 2011
Currently I have the need to build a new Windows XP Guest in VirtualBox. So I figure that I want the latest and greatest so that I wont have to keep upgrading it.
On the OS side, I want it to be pre-installed with SP3, and possibly some or all of the patches
On the Virtual Hardware side I want for the Hard drive to be SATA, not IDE, and for the bridge to be ICH9, not PXII3.
My ultimate goal is for it to be point and shoot, no installing as IDE / PXII3 and convert, but to choose SATA / ICH9 straight from the start. Also want to do it without any other tools like nlite. Starting work today, lets see where this gets us
On the OS side, I want it to be pre-installed with SP3, and possibly some or all of the patches
On the Virtual Hardware side I want for the Hard drive to be SATA, not IDE, and for the bridge to be ICH9, not PXII3.
My ultimate goal is for it to be point and shoot, no installing as IDE / PXII3 and convert, but to choose SATA / ICH9 straight from the start. Also want to do it without any other tools like nlite. Starting work today, lets see where this gets us
Building new Windows XP Guest
Tuesday, July 12, 2011
I was patching my ESXi server today with the VMware Host Update Utility and was thinking about a couple of things. First when I do an update the VMU first downloads a bunch of data to my C:\Documents and Settings\All Users\Application Data\VMware\VMware VI Update\vmw, about 1.2GB of data.
I was thinking that if I wanted to switch back to using a scripting tool then I would have to re-download all of the patches again manually and then script it up. So I wanted the shortcut to see if I really needed to download it again. Inside the MetaData directory, there are four zip files; 4.0 ESX and ESXi, and 4.1 ESX and ESXi. and then in the vibs directory there are a bunch of vibs files ranging from 60 to 130 MB. Those had to be the patches.
Any search on how to open a VIBS file came up empty so I then looked at the Metadata zips again and opened them all up. Inside there is a file called Packages. Once I looked into it I noticed that it was all names of the patches, but was was most interesting was the filename was a .deb file.
So I renamed one of the .vibs file to a .deb copied it over to my Linux server and ran the command ar vx something.deb, and viola, it opened up and showed its contents.
Unfortunately for me, it looks like the contents inside the .deb / .vibs file is not easily compatible the downloaded manual patches, but I learned a little more on how the VMware VMU architecture works.
What I found very interesting was that even though it was for ESXi, these are debian patch packages, so still somewhere in that tiny ESXi Hypervisor, there is some Linux, however small.
Thanks to http://www.g-loaded.eu/2008/01/28/how-to-extract-rpm-or-deb-packages/ on how to extract DEB packages
I was thinking that if I wanted to switch back to using a scripting tool then I would have to re-download all of the patches again manually and then script it up. So I wanted the shortcut to see if I really needed to download it again. Inside the MetaData directory, there are four zip files; 4.0 ESX and ESXi, and 4.1 ESX and ESXi. and then in the vibs directory there are a bunch of vibs files ranging from 60 to 130 MB. Those had to be the patches.
Any search on how to open a VIBS file came up empty so I then looked at the Metadata zips again and opened them all up. Inside there is a file called Packages. Once I looked into it I noticed that it was all names of the patches, but was was most interesting was the filename was a .deb file.
So I renamed one of the .vibs file to a .deb copied it over to my Linux server and ran the command ar vx something.deb, and viola, it opened up and showed its contents.
Unfortunately for me, it looks like the contents inside the .deb / .vibs file is not easily compatible the downloaded manual patches, but I learned a little more on how the VMware VMU architecture works.
What I found very interesting was that even though it was for ESXi, these are debian patch packages, so still somewhere in that tiny ESXi Hypervisor, there is some Linux, however small.
Thanks to http://www.g-loaded.eu/2008/01/28/how-to-extract-rpm-or-deb-packages/ on how to extract DEB packages
Cracking the contents of the VMware VMU vibs files
Cracking the contents of the VMware VMU vibs files
I was patching my ESXi server today with the VMware Host Update Utility and was thinking about a couple of things. First when I do an update the VMU first downloads a bunch of data to my C:\Documents and Settings\All Users\Application Data\VMware\VMware VI Update\vmw, about 1.2GB of data.
I was thinking that if I wanted to switch back to using a scripting tool then I would have to re-download all of the patches again manually and then script it up. So I wanted the shortcut to see if I really needed to download it again. Inside the MetaData directory, there are four zip files; 4.0 ESX and ESXi, and 4.1 ESX and ESXi. and then in the vibs directory there are a bunch of vibs files ranging from 60 to 130 MB. Those had to be the patches.
Any search on how to open a VIBS file came up empty so I then looked at the Metadata zips again and opened them all up. Inside there is a file called Packages. Once I looked into it I noticed that it was all names of the patches, but was was most interesting was the filename was a .deb file.
So I renamed one of the .vibs file to a .deb copied it over to my Linux server and ran the command ar vx something.deb, and viola, it opened up and showed its contents.
Unfortunately for me, it looks like the contents inside the .deb / .vibs file is not easily compatible the downloaded manual patches, but I learned a little more on how the VMware VMU architecture works.
What I found very interesting was that even though it was for ESXi, these are debian patch packages, so still somewhere in that tiny ESXi Hypervisor, there is some Linux, however small.
Thanks to http://www.g-loaded.eu/2008/01/28/how-to-extract-rpm-or-deb-packages/ on how to extract DEB packages
I was thinking that if I wanted to switch back to using a scripting tool then I would have to re-download all of the patches again manually and then script it up. So I wanted the shortcut to see if I really needed to download it again. Inside the MetaData directory, there are four zip files; 4.0 ESX and ESXi, and 4.1 ESX and ESXi. and then in the vibs directory there are a bunch of vibs files ranging from 60 to 130 MB. Those had to be the patches.
Any search on how to open a VIBS file came up empty so I then looked at the Metadata zips again and opened them all up. Inside there is a file called Packages. Once I looked into it I noticed that it was all names of the patches, but was was most interesting was the filename was a .deb file.
So I renamed one of the .vibs file to a .deb copied it over to my Linux server and ran the command ar vx something.deb, and viola, it opened up and showed its contents.
Unfortunately for me, it looks like the contents inside the .deb / .vibs file is not easily compatible the downloaded manual patches, but I learned a little more on how the VMware VMU architecture works.
What I found very interesting was that even though it was for ESXi, these are debian patch packages, so still somewhere in that tiny ESXi Hypervisor, there is some Linux, however small.
Thanks to http://www.g-loaded.eu/2008/01/28/how-to-extract-rpm-or-deb-packages/ on how to extract DEB packages
Labels:
ESX,
ESXi,
Security_Patches,
Vmware,
VMware_tools
Tuesday, November 16, 2010
I have been looking at documents that detail speed improvements with ESX where the partitions are aligned. These documents detail and only deal with SAN paritions:
http://media.netapp.com/documents/tr_3593.pdf
http://www.vmware.com/pdf/vi_performance_tuning.pdf
http://www.vmware.com/pdf/esx3_partition_align.pdf
Now I don't have access to a SAN but wanted to develop and test this with my Direct Attached Storage (DAS) for my ESXi host. Here is the configuration of my ESX host:
Hypervisor OS: ESXi 4.0 Releasebuild 294855
Systemboard: Intel S5000PSL
SCSI Controller: IBM ServeRaid 6M - 128 MB cache
RAID for Testing: RAID5 built with 5 73GB 10K U320 disks.
RAID Stripe Size: 64KB
To do the testing I created two Windows 2003 guests with the following configurations:
OS: Windows 2003 Standard
Service Pack: SP2
Hotfixes: All hotfixes presented over Windows update as of 11/15/2010
Virtual Disk: 12 GB dynamic
Virtual Memory: 512MB
NTFS Format: default (4K)
The only difference is that for SERVERB, the disk was aligned to 64K as described in the documents, and SERVERA had no alignment.
I then tested the servers using iozone.org version 3.53. In order to isolate each server I rebooted the ESX host, then gave it 5 minutes to settle, and started up the Guest target alone with no other guests running with another 5 minutes of settle time before starting the testing.
The tests were done three times for each server and the results were averaged. Then the results for each aggregate server were compared as a percentage difference.
Now the big question, is it worth it. The short answer is yes, I saw an improvement on average of about 10 to 40% per test. There were points where the aligned disk was slower, but if you look at the aggregate of the data there is an improvement. I need to do some more cleanup of the data but I will be posting it soon.
http://media.netapp.com/documents/tr_3593.pdf
http://www.vmware.com/pdf/vi_performance_tuning.pdf
http://www.vmware.com/pdf/esx3_partition_align.pdf
Now I don't have access to a SAN but wanted to develop and test this with my Direct Attached Storage (DAS) for my ESXi host. Here is the configuration of my ESX host:
Hypervisor OS: ESXi 4.0 Releasebuild 294855
Systemboard: Intel S5000PSL
SCSI Controller: IBM ServeRaid 6M - 128 MB cache
RAID for Testing: RAID5 built with 5 73GB 10K U320 disks.
RAID Stripe Size: 64KB
To do the testing I created two Windows 2003 guests with the following configurations:
OS: Windows 2003 Standard
Service Pack: SP2
Hotfixes: All hotfixes presented over Windows update as of 11/15/2010
Virtual Disk: 12 GB dynamic
Virtual Memory: 512MB
NTFS Format: default (4K)
The only difference is that for SERVERB, the disk was aligned to 64K as described in the documents, and SERVERA had no alignment.
I then tested the servers using iozone.org version 3.53. In order to isolate each server I rebooted the ESX host, then gave it 5 minutes to settle, and started up the Guest target alone with no other guests running with another 5 minutes of settle time before starting the testing.
The tests were done three times for each server and the results were averaged. Then the results for each aggregate server were compared as a percentage difference.
Now the big question, is it worth it. The short answer is yes, I saw an improvement on average of about 10 to 40% per test. There were points where the aligned disk was slower, but if you look at the aggregate of the data there is an improvement. I need to do some more cleanup of the data but I will be posting it soon.
Testing to see if aligned partitions on ESX - an Introduction
Testing to see if aligned partitions on ESX - an Introduction
I have been looking at documents that detail speed improvements with ESX where the partitions are aligned. These documents detail and only deal with SAN paritions:
http://media.netapp.com/documents/tr_3593.pdf
http://www.vmware.com/pdf/vi_performance_tuning.pdf
http://www.vmware.com/pdf/esx3_partition_align.pdf
Now I don't have access to a SAN but wanted to develop and test this with my Direct Attached Storage (DAS) for my ESXi host. Here is the configuration of my ESX host:
Hypervisor OS: ESXi 4.0 Releasebuild 294855
Systemboard: Intel S5000PSL
SCSI Controller: IBM ServeRaid 6M - 128 MB cache
RAID for Testing: RAID5 built with 5 73GB 10K U320 disks.
RAID Stripe Size: 64KB
To do the testing I created two Windows 2003 guests with the following configurations:
OS: Windows 2003 Standard
Service Pack: SP2
Hotfixes: All hotfixes presented over Windows update as of 11/15/2010
Virtual Disk: 12 GB dynamic
Virtual Memory: 512MB
NTFS Format: default (4K)
The only difference is that for SERVERB, the disk was aligned to 64K as described in the documents, and SERVERA had no alignment.
I then tested the servers using iozone.org version 3.53. In order to isolate each server I rebooted the ESX host, then gave it 5 minutes to settle, and started up the Guest target alone with no other guests running with another 5 minutes of settle time before starting the testing.
The tests were done three times for each server and the results were averaged. Then the results for each aggregate server were compared as a percentage difference.
Now the big question, is it worth it. The short answer is yes, I saw an improvement on average of about 10 to 40% per test. There were points where the aligned disk was slower, but if you look at the aggregate of the data there is an improvement. I need to do some more cleanup of the data but I will be posting it soon.
http://media.netapp.com/documents/tr_3593.pdf
http://www.vmware.com/pdf/vi_performance_tuning.pdf
http://www.vmware.com/pdf/esx3_partition_align.pdf
Now I don't have access to a SAN but wanted to develop and test this with my Direct Attached Storage (DAS) for my ESXi host. Here is the configuration of my ESX host:
Hypervisor OS: ESXi 4.0 Releasebuild 294855
Systemboard: Intel S5000PSL
SCSI Controller: IBM ServeRaid 6M - 128 MB cache
RAID for Testing: RAID5 built with 5 73GB 10K U320 disks.
RAID Stripe Size: 64KB
To do the testing I created two Windows 2003 guests with the following configurations:
OS: Windows 2003 Standard
Service Pack: SP2
Hotfixes: All hotfixes presented over Windows update as of 11/15/2010
Virtual Disk: 12 GB dynamic
Virtual Memory: 512MB
NTFS Format: default (4K)
The only difference is that for SERVERB, the disk was aligned to 64K as described in the documents, and SERVERA had no alignment.
I then tested the servers using iozone.org version 3.53. In order to isolate each server I rebooted the ESX host, then gave it 5 minutes to settle, and started up the Guest target alone with no other guests running with another 5 minutes of settle time before starting the testing.
The tests were done three times for each server and the results were averaged. Then the results for each aggregate server were compared as a percentage difference.
Now the big question, is it worth it. The short answer is yes, I saw an improvement on average of about 10 to 40% per test. There were points where the aligned disk was slower, but if you look at the aggregate of the data there is an improvement. I need to do some more cleanup of the data but I will be posting it soon.
Friday, October 29, 2010
OK, I know the title is a little long, but it is descriptive. Yesterday I stumbled across the Zabbix appliance and wanted to see what it could do.
Well I am still checking out Zabbix, but I wanted to see if I could run the Zabbix appliance using VirtualBox's native disk format VDI instead of the originally provided VMDK virtual disk.
First off adding the VMDK into VirtualBox is easy as cake
Now for the fun part; converting the file to VDI
Use the following command:
vboxmanage.exe clonehd\file.vmdk zabbix.vdi --format vdi
Now you just go to the Virtual Media Manager and attach the VDI and away you go. The one thing that I did learn is that you need to convert the VMDK to a VDI before you start up the machine the first time. So if you did the test with at the top of the article, just delete it and pull a new copy of the VMDK from the archive you download.
Well I am still checking out Zabbix, but I wanted to see if I could run the Zabbix appliance using VirtualBox's native disk format VDI instead of the originally provided VMDK virtual disk.
First off adding the VMDK into VirtualBox is easy as cake
- Download the appliance from www.zabbix.com
- Extract the VMDK and VMX files.
- By looking at the VMX file, you notice the parameters that are pertinent to create
- OS: Linux
- Version: OpenSUSE
- Memory: 512MB
- Storage Controller: SATA
- Network Attached Adapter: Bridged
- Adapter type Intel PRO/1000 MT
- Copy the VMDK to where you keep your VDI files, go to the Virtual Media Manager and add the disk to the list, then attach the disk to the SATA adapter and away you go.
Now for the fun part; converting the file to VDI
Use the following command:
vboxmanage.exe clonehd
Now you just go to the Virtual Media Manager and attach the VDI and away you go. The one thing that I did learn is that you need to convert the VMDK to a VDI before you start up the machine the first time. So if you did the test with at the top of the article, just delete it and pull a new copy of the VMDK from the archive you download.
Installng Zabbix appliance into Virtualbox with VMDK to VDI conversion
Installng Zabbix appliance into Virtualbox with VMDK to VDI conversion
OK, I know the title is a little long, but it is descriptive. Yesterday I stumbled across the Zabbix appliance and wanted to see what it could do.
Well I am still checking out Zabbix, but I wanted to see if I could run the Zabbix appliance using VirtualBox's native disk format VDI instead of the originally provided VMDK virtual disk.
First off adding the VMDK into VirtualBox is easy as cake
Now for the fun part; converting the file to VDI
Use the following command:
vboxmanage.exe clonehd\file.vmdk zabbix.vdi --format vdi
Now you just go to the Virtual Media Manager and attach the VDI and away you go. The one thing that I did learn is that you need to convert the VMDK to a VDI before you start up the machine the first time. So if you did the test with at the top of the article, just delete it and pull a new copy of the VMDK from the archive you download.
Well I am still checking out Zabbix, but I wanted to see if I could run the Zabbix appliance using VirtualBox's native disk format VDI instead of the originally provided VMDK virtual disk.
First off adding the VMDK into VirtualBox is easy as cake
- Download the appliance from www.zabbix.com
- Extract the VMDK and VMX files.
- By looking at the VMX file, you notice the parameters that are pertinent to create
- OS: Linux
- Version: OpenSUSE
- Memory: 512MB
- Storage Controller: SATA
- Network Attached Adapter: Bridged
- Adapter type Intel PRO/1000 MT
- Copy the VMDK to where you keep your VDI files, go to the Virtual Media Manager and add the disk to the list, then attach the disk to the SATA adapter and away you go.
Now for the fun part; converting the file to VDI
Use the following command:
vboxmanage.exe clonehd
Now you just go to the Virtual Media Manager and attach the VDI and away you go. The one thing that I did learn is that you need to convert the VMDK to a VDI before you start up the machine the first time. So if you did the test with at the top of the article, just delete it and pull a new copy of the VMDK from the archive you download.
Labels:
VboxManage,
VirtualBox,
VMDK_Converson,
Zabbix
Thursday, August 19, 2010
After a long and hard trial process, I have increased the speed of my RAID5 array on an IBM ServeRAID-6M to a very acceptable level
Before I worked on this, guests on a 5 disk RAID5 were getting 5MB/s and after-wards I am getting the following Speeds
Before I worked on this, guests on a 5 disk RAID5 were getting 5MB/s and after-wards I am getting the following Speeds
- Windows 2003, 12GB, Read tests with HD_speed are 92 MB/s
- Ubuntu 9.04 server, 8GB Read+Write tests, using dd are 32 MB/s
- Test the READ and WRITE speeds of each disk by itself. I had 1 U320 disk that was reading fine at 60 MB/s but crippled down and would only write at 5 MB/s
- Have the same interface; one disk was a U160, and it could only read at 50 MB/s
- Do not include any disks that are considerably slower than others, it will slow down the whole array.
- Not sure this is true with all Array controllers, but make sure that both the disks and the array are set to write back, not write through. Also the setting for the disks can only be change while they are not in an array, and cannot be change once they are built into an array
- When building the array use the largest stripe size. The ServeRAID-6M goes up to 64K
- When formatting the VMFS Datastore choose a larger block size like 4MB or 8MB.
Success in improving RAID speeds
Success in improving RAID speeds
After a long and hard trial process, I have increased the speed of my RAID5 array on an IBM ServeRAID-6M to a very acceptable level
Before I worked on this, guests on a 5 disk RAID5 were getting 5MB/s and after-wards I am getting the following Speeds
Before I worked on this, guests on a 5 disk RAID5 were getting 5MB/s and after-wards I am getting the following Speeds
- Windows 2003, 12GB, Read tests with HD_speed are 92 MB/s
- Ubuntu 9.04 server, 8GB Read+Write tests, using dd are 32 MB/s
- Test the READ and WRITE speeds of each disk by itself. I had 1 U320 disk that was reading fine at 60 MB/s but crippled down and would only write at 5 MB/s
- Have the same interface; one disk was a U160, and it could only read at 50 MB/s
- Do not include any disks that are considerably slower than others, it will slow down the whole array.
- Not sure this is true with all Array controllers, but make sure that both the disks and the array are set to write back, not write through. Also the setting for the disks can only be change while they are not in an array, and cannot be change once they are built into an array
- When building the array use the largest stripe size. The ServeRAID-6M goes up to 64K
- When formatting the VMFS Datastore choose a larger block size like 4MB or 8MB.
Tuesday, August 17, 2010
I have been working with a test version of ESX 4i, 4.0 U2 on a USB stick. I installed it the USB stick like I documented on my August 4th post. I like it but seeing some problems. For some reason it is slowing down my system. When I benchmark my ESX host it is only getting 5MB/s on disk access for the guests.
Now these guests are either on a single SCSI U320 disk attached to a ServeRaid-6M, or in a RAID 5 config on that same card, and they are behaving poorly.
I believe that somehow the ESX hypervisor is writing to the USB stick and slowing the whole system down. So I am going back to an installable version on a SCSI disk
Now these guests are either on a single SCSI U320 disk attached to a ServeRaid-6M, or in a RAID 5 config on that same card, and they are behaving poorly.
I believe that somehow the ESX hypervisor is writing to the USB stick and slowing the whole system down. So I am going back to an installable version on a SCSI disk
ESXi on USB - Part II
ESXi on USB - Part II
I have been working with a test version of ESX 4i, 4.0 U2 on a USB stick. I installed it the USB stick like I documented on my August 4th post. I like it but seeing some problems. For some reason it is slowing down my system. When I benchmark my ESX host it is only getting 5MB/s on disk access for the guests.
Now these guests are either on a single SCSI U320 disk attached to a ServeRaid-6M, or in a RAID 5 config on that same card, and they are behaving poorly.
I believe that somehow the ESX hypervisor is writing to the USB stick and slowing the whole system down. So I am going back to an installable version on a SCSI disk
Now these guests are either on a single SCSI U320 disk attached to a ServeRaid-6M, or in a RAID 5 config on that same card, and they are behaving poorly.
I believe that somehow the ESX hypervisor is writing to the USB stick and slowing the whole system down. So I am going back to an installable version on a SCSI disk
Wednesday, August 4, 2010
I found this great little tutorial on how to install ESXi onto a USB stick. I know that there are a bunch of ways, but this was fast, simple, and can be done from windows with tools that I already have: WinRAR and WinImage
http://www.vladan.fr/how-to-install-esxi-40-on-usb-memory-key
Give it a shot.
UPDATE: I use this on a dedicated system so I am not concerned with data loss, but from my tests with 4.0 U2 the hypervisor will grab a partition and format it with VMFS.
http://www.vladan.fr/how-to-install-esxi-40-on-usb-memory-key
Give it a shot.
UPDATE: I use this on a dedicated system so I am not concerned with data loss, but from my tests with 4.0 U2 the hypervisor will grab a partition and format it with VMFS.
ESXi on USB
ESXi on USB
I found this great little tutorial on how to install ESXi onto a USB stick. I know that there are a bunch of ways, but this was fast, simple, and can be done from windows with tools that I already have: WinRAR and WinImage
http://www.vladan.fr/how-to-install-esxi-40-on-usb-memory-key
Give it a shot.
UPDATE: I use this on a dedicated system so I am not concerned with data loss, but from my tests with 4.0 U2 the hypervisor will grab a partition and format it with VMFS.
http://www.vladan.fr/how-to-install-esxi-40-on-usb-memory-key
Give it a shot.
UPDATE: I use this on a dedicated system so I am not concerned with data loss, but from my tests with 4.0 U2 the hypervisor will grab a partition and format it with VMFS.
Subscribe to:
Posts (Atom)