New Domain Name.
Point your feeds to the new address! http://feeds.feedburner.com/RogerLundsItVmware/VirutalizationBlog
Roger Lund
Category:
News
We paid for Vmware support and I had an event. Nothing big but what the heck we paid for support. I created a ticket through the support portal and got a call back the same day. I talked with the tech and decided to postpone the call until the following week since it was Friday and it was nothing big. The next week the tech called me when he had promised. He remote desk-toped in and we setup a putty session. I watched him use the following commands. Some are basic but some are new it me.
cd etc
less hosts
cd var
less hostd.log
date
ps -ef | grep hostd
ps -ef | grep vpxa
config.xml -u
service vmware status
cat vmkernel
vdf -h
vmware -v
esxupdate
top
view vmkernel
ps -lf -C cimserver
I hope these commands are helpful.
Ben T.
Cody at Professional VMware has a nice write up on VMDirectPath and Paravirutal SCSI, titled : VMDirectPath? Paravirtual SCSI? – vSphere VM Options and You!
“This post comes because I am just as confused as the rest of you when it comes to the options available in some of these new vSphere interfaces. I figure it best to take a look at some of the options, and figure when it is best to move away from defaults and start tuning things just right.
This one is another post that originated from the great Mr. @rogerlund.
Specifically Mr. Lund asked me to answer the following three questions: “Which applications would be good for Paravirtual SCSI? Which for VMDirectPath? And; How do we choose?”
“
Paravirtual SCSI (PVSCSI)
Part (most) of the when & why for PVSCSI is provided in the quote from above. You use PVSCSI when you need a high performance virtual storage adapter. What does this mean? It means you will not use this for your AD server, or print server. You also generally will not use this on local storage, or DAS (Direct Attached Storage).
When would you use it? Well, I’m glad you asked. Remember that database that you were not virtualizing, because of it’s high IO requirement? That graphics rendering app for marketing that let the magic blue smoke out of your last SAN array? These are good candidates for PVSCSI.
VMDirectPath
This here, is where we answer the when and why, for VMDirectPath. VMDirectPath is quite a bit different than PVSCSI, but no less cool. It allows VMs to directly access PCI(e) devices. Up to two per VM. Now why would you want to do that? Wasn’t hardware abstraction one of the beauties of virtualizing your environment? Remember that security appliance? The one that corporate security insists be physical as it has this oddball entropy card? That goes here.
”
I only quoted a couple key parts, please refer to the full post to read the full article http://professionalvmware.com/2009/08/17/vmdirectpath-paravirtual-scsi-vsphere-vm-options-and-you/
Thanks to Cody for writing this up!
I did want to cover the steps on setting both of these up. The following steps are taken from the VMware vSphere Online Library.
“
Add a Paravirtualized SCSI Adapter
- Right-click on the virtual machine and select Edit Settings.
- Click Add.
- Select SCSI Device and click Next.
- Select a SCSI device.
- Select an unused Virtual Device Node.
- Click Next.
- Review your selections and click Finish.
- A new SCSI device and a new SCSI controller are created.
- Select the new SCSI controller and click Change Type.
- Select VMware Paravirtual and click OK.
“
PCI devices connected to a host can be marked as available for passthrough from the Hardware Advanced Settings in the Configuration tab for the host.
To use VMDirectPath, the host must have Intel® Virtualization Technology for Directed I/O (VT-d) or AMD I/O Virtualization Technology (IOMMU) enabled in the BIOS. In order to add PCI devices to a virtual machine, the devices must be connected to the host and marked as available for passthrough. In addition, PCI devices can be added only to virtual machines with hardware version 7.
- Select the virtual machine from the inventory panel and click Virtual Machine > Edit Settings.
- On the Hardware tab, click Add.
- In the Add Hardware wizard, select PCI Device and click Next.
- Select the passthrough device to connect to the virtual machine from the drop-down list and click Next.
- Click Finish.
“
With Cody’s Write up, and the above directions, we should be set, have fun playing!
Roger L.
Yes, I am very behind on reading KB Articles.
Here are a few that I saw, and wanted to link, I have the symptoms here, you’ll have to click the links for resolutions.
Issues when vCenter Server and License Server are installed on ESX 3.x hosted virtual machines
KB Article: 1011085 Updated: Jun 23, 2009 Symptoms
If you have vCenter Server and License Server installed on a virtual machine on an ESX 3.x host, you may experience the following symptoms:
- You cannot power on the virtual machine
- You see the error not enough licenses to perform the operation if you:
- Try to power on the virtual machine
- Reboot the virtual machine
- Try to add new licenses to the server
Note: This error displays even though you correctly installed a server-based (centralized) license file.- You cannot change the ESX status from Unlicensed. All other options are unavailable.
- You cannot use vCenter features such as VMotion, VMware High Availability, and Distributed Resource Scheduling
The tg3 network driver fails to load during ESX boot
KB Article: 1006601 Updated: Jun 19, 2009 Symptoms
You may experience these symptoms:
- When migrating a host to a new network, the onboard NICs stop working
- Broadcomm tg3.0 kernel driver is not loading at boot
- You can manually load the tg3 driver successfully
Issues when vCenter Server and License Server are installed on ESX 3.x hosted virtual machines
Symptoms
If you have vCenter Server and License Server installed on a virtual machine on an ESX 3.x host, you may experience the following symptoms:
- You cannot power on the virtual machine
- You see the error not enough licenses to perform the operation if you:
- Try to power on the virtual machine
- Reboot the virtual machine
- Try to add new licenses to the server
Note: This error displays even though you correctly installed a server-based (centralized) license file.- You cannot change the ESX status from Unlicensed. All other options are unavailable.
- You cannot use vCenter features such as VMotion, VMware High Availability, and Distributed Resource Scheduling
VirtualCenter Server with multiple IP addresses changes the IP address while the services are running
Symptoms
- The VirtualCenter Server has multiple IP addresses and changes the IP address on which the services run
- The ESX hosts disconnect from VirtualCenter
- You see the message:
These ESX are currently being managed by another VirtualCenter (VC) and do you want to continue.
The "other" VirtualCenter (VC) is using the original IP address of the server.
Cannot see a line item for add-ons when viewing licenses on the vSphere Client licensing page
Symptoms
- When you view your purchase on the vSphere Client licensing page, a separate product line item for add-ons is not displayed
- Only vSphere 4.0 Standard license appears even though you have purchased, for example, a vSphere 4.0 Standard with VMotion license
Unable to start vCenter 4 after upgrade from 2.0.2
Symptoms
- After successful upgrade from VirtualCenter 2.0.2 to vCenter Server, the vCenter service does not start
- vpxd logs show:
[fusion_builder_container hundred_percent=”yes” overflow=”visible”][fusion_builder_row][fusion_builder_column type=”1_1″ background_position=”left top” background_color=”” border_size=”” border_color=”” border_style=”solid” spacing=”yes” background_image=”” background_repeat=”no-repeat” padding=”” margin_top=”0px” margin_bottom=”0px” class=”” id=”” animation_type=”” animation_speed=”0.3″ animation_direction=”left” hide_on_mobile=”no” center_content=”no” min_height=”none”][VpxUtil] Failed to deserialize object using SOAP deserializer: Invalid argument : version
[VpxdInvtVm] Failed to load vm summary: Not initialized: boolean disableSnapshotsSupported
[VpxdMain] Failed to initialize: Not initialized: boolean disableSnapshotsSupported
Failed to initialize VMware VirtualCenter. Shutting down…
Forcing shutdown of VMware VirtualCenter now
vSphere Host Update Utility not available for download to upgrade standalone ESX Server 3.x to ESX 4
Purpose
If you do not have access to vCenter 4.0 then the vSphere Host Update Utility cannot be installed. The installer is only available with the installation of vCenter 4.0 and is currently not available for download
Cannot deploy a virtual machine from a template after changing the NFS datastore IP address
Details
- You cannot deploy a virtual machine from a template after changing the IP address of an NFS datastore
- You see the error:
unknown device ID. - You see the error:
Invalid configuration on device 1 . The specified key name identifier already exists. - When converting the template to a virtual machine, you see the error:
host or cluster is not valid. - If you delete the NFS datastore from all ESX hosts in the cluster and then add it back using the VirtualCenter Server, a (1) is appended to the original datastore name.
Cannot log in to an ESX 4 host with vSphere Web Access
Symptoms
- You cannot log in to an ESX 4.0 host using vSphere Web Access
- The vSphere Web Access browser does not open
- The browser shows the error:
503 Service unavailable
Today, I had a fellow VI admin, tell me that my blog was a #1 hit on Google, regarding a VMware KB.
Thought I should keep it up.
Roger L.
[/fusion_builder_column][/fusion_builder_row][/fusion_builder_container]
Simon Long wrote a blog post on NPIV support in vSphere titled : NPIV support in VMware ESX4
“Whilst revising for the VCP4 Beta Exam and also replying to a thread on the VMTN Forum, I’ve come across a couple of instances where there is a lack of “using NPIV in VMware ESX 4″ information. The only good post I can find is Jason Boche’s post; N_Port ID Virtualization (NPIV) and VMware Virtual Infrastructure, but his post is written and tested using ESX3.5. So I have decided to find out as much information as I can and post it here.
What are the Advantages of using NPIV?
- Standard storage management methodology across physical and virtual servers.
- Portability of access privileges during VM migration.
- Fabric performance, as NPIV provides quality of service (QoS) and prioritization for ensured VM-level bandwidth assignment.
- Auditable data security due to zoning (one server, one zone).
“
Read the full blog post at http://www.simonlong.co.uk/blog/2009/07/21/npiv-support-in-vmware-esx4/
I wanted to post this as very useful info.
Thanks to Simon.
I have been behind on my blog reading, so here are a few posts that I thought were worth the read.
VMware vSphere Thin Provisioning
by http://www.virtualpro.co.uk/
“I’ve recently been evaluating some of the new features in VMware vSphere to see what use they would be to my current employer. One of the areas that I touched upon in my “what’s new in vSphere Storage” blog post was thin provisioning. I wanted to come back and cover this particular topic in more detail as it’s a key feature and it’s available throughout all versions of vSphere so I’m sure everyone will be interested in it.”
How much RAM for an ESX server – a.k.a. Lego Blocks
by http://technodrone.blogspot.com/
“
I started to read the sample chapters that Scott Lowe released from his upcoming book, and one of the parts were about the subject of scaling up vs. scaling out.
A slight bit more of an explanation as to what I mean by this. Should I buy bigger more monstrous servers, or a greater number of smaller servers? “
FT Problem Decoder Chart
“Receiving errors while trying to configure FT (Fault Tolerance) on a VM and stumped as to the reason why? This may help.”
Not All FT Compatible CPUs Are Created Equal
“Hopefully you are aware that to enable VMware vSphere’s FT (Fault Tolerance), you need FT compatible CPUs from Intel or AMD. VMware KB article 1008027 Processors and guest operating systems that support VMware Fault Tolerance outlines both the Intel and AMD CPU requirements to use FT. I had read this article months ago and on that basis I purchased FT compatible AMD Opteron 2356 Barcelona Quad Core processor upgrades for the HP DL385 G2 servers in my lab.”
Max amount of VMs per VMFS volume
by http://www.yellow-bricks.com
“Two weeks ago I discussed how to determine the correct LUN/VMFS size. In short it boils down to the following formula:
round((maxVMs * avgSize) + 20% )
So in other words, the max amount of virtual machines per volume multiplied by the average size of a virtual machine plus 20% for snaps and .vswp rounded up. (As pointed out in the comments if you have VMs with high amounts of memory you will need to adjust the % accordingly.) This should be your default VMFS size. Now a question that was asked in one of the comments, which I already expected, was “how do I determine what the maximum amount of VMs per volume is?”. There’s an excellent white paper on this topic. Of course there’s more than meets the eye but based on this white paper and especially the following table I decided to give it a shot:”
Xen : Bridging on bonded & trunked interfaces
by http://unixfoo.blogspot.com
"When you deploy some critical stuffs like Oracle DB, Oracle RAC DB etc on the Xen or Oracle virtual machine, you’ll have to use bonded network interfaces and the vlan trunks. By default, Xen and Oracle VM doesn’t support the xen bridges created on bonded-vlaned-interfaces. The below document gives you a clear idea on how to create and configure xen bridges on bonded and vlan trunked interfaces.”
Central MN IT User Group , presents a VMware vSphere & SharePoint 101 User Group Meeting.
by Roger Lund
written by Roger Lund
Central MN IT User Group , presents a VMware vSphere & SharePoint 101 User Group Meeting.
Brought to you by http://www.educait.net
8/13/2009
Central MN IT User Group Meeting agenda
1:30 – 2:30 VMware 4.0 overview, new features.
2:30 – 2:40 break
2:40 – 3:10 Case Study
3:10 – 3:20 break
3:20 – 4:10 SharePoint overview
4:10 – 4:30 Open Discussion
Registration:
http://centralmnit2-emailinvite.eventbrite.com/
We have the Mississippi room, which fits well over 100 people.
Please pass this on. Remember, this is a free event.
VI / vCenter cannot connect to Host, Service mgmt-vmware restart may not restart hostd
by Roger Lund
written by Roger Lund
Yesterday, I had a problem after I shutdown a lagcy San, that after I rescaned my HBA’s, ( which timed out), I lost connection to the host.
I tried restarting the normal services, in console,
# service vmware-vpxa restart
#service mgmt-vmware restart
but I would get,
Stopping VMware ESX Server Management services:
VMware ESX Server Host Agent Watchdog [fusion_builder_container hundred_percent=”yes” overflow=”visible”][fusion_builder_row][fusion_builder_column type=”1_1″ background_position=”left top” background_color=”” border_size=”” border_color=”” border_style=”solid” spacing=”yes” background_image=”” background_repeat=”no-repeat” padding=”” margin_top=”0px” margin_bottom=”0px” class=”” id=”” animation_type=”” animation_speed=”0.3″ animation_direction=”left” hide_on_mobile=”no” center_content=”no” min_height=”none”][FAILED]
I found the following post , http://communities.vmware.com/message/1157237 and KB article, http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1005566
Both which have you kill the hostd, process, then rename the pid, file. Here is the contents of the KB.
"
Service mgmt-vmware restart may not restart hostd
Symptoms
- You are using the command service mgmt-vmware restart but it does not finish restarting hostd.
- The script gets stuck when stopping the service.
The SSH session to the ESX host becomes unresponsive.
hostd does not restart
Resolution
You must manually stop the stuck service and restart it.
To stop the service and restart it:
Log in as root to the ESX host command-line via the physical console or via KVM connection.
- Navigate to the /var/run/vmware directory:
# cd /var/run/vmwareRun the following command to list the files vmware-hostd.PID and watchdog-hostd.PID:
# ls -l vmware-hostd.PID watchdog-hostd.PID- Determine the Process ID (PID) management service. View the contents of the vmware-hostd.PID file:
# cat vmware-hostd.PID
For example:
[root@vmware]# cat vmware-hostd.PID
1191[root@vmware]#Use the resulting PID to kill the process.
Caution: Use the kill -9 command with care. It kills the process of the supplied PID without exception or confirmation.
# kill -9 <PID>
In this example you run kill -9 1191.- Delete the vmware-hostd.PID and watchdog-hostd.PID files:
# rm vmware-hostd.PID watchdog-hostd.PID- Restart the management service.
# service mgmt-vmware restart”
This seemed to fix it, and without any downtime.
Roger L
[/fusion_builder_column][/fusion_builder_row][/fusion_builder_container]
I get this question often, in fact I had this question at our first local user group.
My Answer was 500 GB, or very close to it, here is a post by http://www.yellow-bricks.com titled : VMFS/LUN size?
“A question that pops up on the VMTN Community once every day is what size VMFS datastore should I create? The answer always varies, one says “500Gb” the other says “1TB”. Now the real answer should be, it depends.
Most companies can use a simple formula in my opinion. First you should answer these questions:
- What’s the maximum amount of VMs you’ve set for a VMFS volume?
- What’s the average size of a VM in your environment? (First remove the really large VM’s that typically get an RDM.)
If you don’t know what the maximum amount of VMs should be just use a safe number, anywhere between 10 and 15. Here’s the formula I always use:
round((maxVMs * avgSize) + 20% )
I usually use increments of 25GB. This is where the round comes in to play. If you end up with 380GB round it up to 400GB and if you end up with 321GB round it up to 325GB. Let’s assume your average VM size is 30GB and your max amount of VMs per VMFS volume is 10:
(10*30) + 60 =360
360 rounded up –> 375GB”
Source Post http://www.yellow-bricks.com/2009/06/23/vmfslun-size/
Seems to me a good method of scaling your datastore sizes.
I also saw this NetApp Article, referring to the above post.
What’s the right LUN size for a VMFS datastore?
“
I can’t tell you how often we are asked this question and the answer has many variables including the amount of aggregated I/O of the servers being virtualized onto the datastore, the storage protocol used in connecting to the datastore, etc.
I believe Duncan Epping at Yellow Bricks has provided us with a simple answer to this traditionally difficult question. Hopefully you’re very familiar with Duncan, he’s an active contributor to the VMTN and his site is one of the top VMware blog sites that is always brimming with the type of information that can significantly improve your VMware deployment.
I compliment Duncan on the way he phrased his guidance. I believe we may want to consider rewording the recommendations currently in TR-3428 (VI3) and the upcoming TR-3749 (vSphere). BTW – the current guidance in both of these documents is a LUN ranging in size from 300 GB – 700 GB.
Go check out Duncan’s post, it contains very solid guidance to this long running question, and may I suggest you drop him a note for his thoughts on sizing a NFS datastore. “
Source Post http://blogs.netapp.com/virtualstorageguy/2009/06/whats-the-right-lun-size-for-a-vmfs-datastore.html
Both Good Reads, enjoy.
ESX host continually disconnects from vCenter following ESX 4 Upgrade
by Roger Lund
written by Roger Lund
I ran into this after I upgraded my vCenter to 4.0, so I thought I would share. KB 1011647
- “KB Article: 1011647
- Updated: Jun 12, 2009
Symptoms
- You upgraded your VirtualCenter instance to vCenter 4.x
- You upgraded ESX hosts to ESX 4.x
- Your ESX disconnect within a couple of seconds of being connected added to vCenter
- Upon reconnect, host disconnects again within 30 seconds
- ESX 3.5 hosts remain connected in VirtualCenter
- ESX not responding after upgrade to ESX 4
Resolution
This issue occurs if you have recently upgraded to ESX 4.0 and vCenter 4.0.
This can occur if vCenter incorrectly pushed the loopback IP address to the vCenter agent configuration. This can be verified by opening the /etc/opt/vmware/vpxa/vpxa.cfg on the ESX 4 host and checking to see if <serverIp> contains 127.0.0.1.
To resolve the issue, in vCenter:
- Click Administration > vCenter Server Settings.
- Click Runtime Settings.
- Under Managed IP address, type the IP of the vCenter Server.
- Connect the ESX host. It now remains connected. “
Source Post. http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1011647
Worked for me, hope this helps someone in the future.
