A few weeks a go I was moving a customer from an old set of ESX servers (not HA clustered) to a new infrastructure of Clustered ESX hosts. After building, testing and verifying the hosts we started moving the VM’s. It became apparent after a little while there were some resource issues. After just a few VM’s were moved an alert appeared that we could not start any new machines. I start looking at the cluster and there is plenty of extra Memory and CPU. Still nothing will start.
I say to myself, “Self, we have read about this before.” I thought back to this HA Deep Dive article by Duncan Epping.
Lets check the HA slots! (on a side note, if you use HA and have never read the Deep Dive, go do it now!)
As you can see here the slot size is rather giant. We have the largest CPU and Memory reservation plus some overhead (for simplicity) and that blows the size of the slot way up. I didn’t set the reservation, but surely they were there. 8GB of reserved memory. 4000MHz of CPU. Ouch. Where did that come from? It followed the VM from the old host to the new one. One of the reasons I was there was to setup a new cluster since the older ones were performing so slow on the local storage. It seems like someone tried to help some critical VM’s along the way by adding the reservations. I removed the reservations and had plenty of slots as you see below.
Yeah! I was able to power on another VM!
The new cluster blew away the old one. Went from older Xeon’s to 6 core Nehalem’s, from local disks to 48 disks of Equallogic Storage. The reservation was no longer needed.
Lessons:
1. Be careful with reservations, it can impact your failover capacity.
2. Reservations set on the machine will follow it to a new host.
*Disclaimer I work for a Xsigo Partner, but I also think it is cool stuff. I do not get any compensation from any vendor for any blog post. My opinions are my own, unless I am copying them from someone.
I recently had someone on twitter ask if there were more Xsigo resources out there for research. He commented that it was really cool, but seemingly hard to find information. I tried to find good information that had some technical content (not just a press release). So I will post some links here and as I come across more I will update this post.
Special thanks to Kevin Miller (@kevin_miller), for making sure I didn’t burn up anything and running out to Fry’s to get a new CPU when the orginal we ordered turned out to be not compatible.
List of the material used in final Version:
Intel DH57JG – Motherboard
Intel i5-661 Core Duo – Processor (special note: DO NOT get the i5-655k unlocked CPU in the picture, it was not compatible with any intel Motherboard)
Lian Li MINI ITX Case (6x9x12 in dimensions) supports Full size PSU
Rosewill silent PSU
Intel dual port PCIe 16x Gbe NIC
Stock intel heatsink and fan (Nexus fan in the picture was very cool, just too big to fit anything into the PCIe port.
2x Intel 80GB SSD
2x4GB OCZ Memory Kit
The server built amazingly fast using PXE boot and installing an ESXi image. Only thing that didn’t work that I wish it did was the on board NIC, 2 GigE ports will be more than enough for our purpose. This is not a lab machine but will run all the components I need to deploy a solution in an automated fashion. I really like VMware Fusion but if I try to run Windows 7, a Linux Server and a Windows 2003 server at the same time my awesome Mac Book Pro becomes useless. This is a lightweight and easy way to bring all of these components together for relatively low cost and if it can save half a day on an install it will pay for itself pretty fast.
I am running:
Windows 2003 ( a vCenter Template)
UDA20 – Ultimate Deployment Appliance that I fought with for a while and finally have it running reliably.
vSphere Mangement Assistant
As of this post working on building:
Windows 7 VM (for Powershell and other tools)
Maybe a free NFS/CIFS server for some easy file shares.
Linux Server (just in case I need it)
Now for some pics:
Before:
SSDs Mounted
Two 80 GB SSD’s are mounted in this space.
Everything going into the Case
Size relative to my hand
I do not have NBA player sized hands, hopefully this shows just how small the case is. This solution is a little more pricey for just a lab machine but if it travels I do not need a SATA disk failing because the server had a bumpy ride.
I usually don’t just make announcements on this site. There are plenty of good news sites that let you know what is new and whatnot. I don’t remember if I saw this but I am behind on my reader feeds so forgive me if everyone already checked this out. The View Open Client is an open client (hence the name) to connect to VMware View Managed Desktop deployments if your OS is Mac or Linux and this is very good news. The java plugin or whatever you call it from the View web manager is annoying to me mainly because the java security is pesky. I know you can just change the settings in Java but I have had in the past where an update from Apple kills my previous settings.
All this to say someone today had never heard of the client and I had show him what was up. He was very happy to be introduced to the View Open client. So I thought it would be cool to spread the word a little more.
Last Friday night I received an email from John Troyer at VMware inviting me to the vExpert 2010 program. It is an honor to be included with such a great list of people. I never know how to respond to be complimented. That is why I like high fives. It is like saying good job, but the only expected response is to high five back. No awkward thank you, or speeches needed. Just don’t leave me hangin’ and everything is understood.
I am extremely excited to get the VMware vExpert 2010, but since this is the internets, no one is here to high five.
Sometimes you may be required to run your vCenter server that has two network interfaces. One in the network it can be reached for remote desktop access and the other where it has access to the ESX servers in order to manage the VMware hosts. This is sort of a hybrid model of an isolated management network. Where only one host can reach the management ports. One thing to think about in this model is Update Manager by default will not like it. Everything may look ok, but trying to scan a host will fail. Luckily though it is an easy fix.
In the update manager configuration tab change the ip in the picture to the IP accessible by the ESX servers. Then remember to restart the Update Manager services. Now go back and run the ESX scan/stage/remediation.
In college I often would be invited to a get together that could often include the letters BYOB, Bring Your Own Beer. Sometimes a cookout would be BYOM, Bring Your Own Meat (or meat alternative for the vegetarians). So today I want to leverage this to push my new acronym B.Y.O.P. Bring Your Own Pod. Lately I have been seeing people talk about Vblocks. If I can venture a succinct definition a Vblock is a pre-configured set of Cisco, EMC and VMware products tested by super smart people, approved by these people to work together, then supported by these organizations as a single entity. Your reseller/solutions provider really should already be doing this very thing for you. You may choose to buy just the network piece, or the hypervisor but your partner should be able to verify a solution to work from end to end and provide unified support.
So You can’t call it BYOPCVCEP
Why not Vblock? This might get me blacklisted by the Elders of the vDiva council, but VCE doesn’t exist to make your life in the datacenter easier, they exist to sell you more VMware, Cisco and EMC. Vblock for sure simplifies your buying experience. I believe they are all great products and may very well do just what you need. Without competition though the only winner is VCE. Do not by forced into a box by the giant vendors. Find someone that can help determine your end goal, provide you vendor neutral analysis of the building blocks needed to achieve your end goal. Then provide the correct vendors and unified support to Build Your Own Pod.
So What is the Alternative Vblock
Originally I was going to draw up a sweet solution of 3par, Xsigo and Dell R610′s and say, “Hey everyone! This is some cool stuff. Try to quiet the overwhelmingly loud voice calling from VCE and give this Alternative Vblock a try.” As I thought more and more about it I think doing that is contrary to my main point. I would like more to provide the discussion points or some possible products among others that can be used to Build Your Own Pod. I am a firm believer in getting what is right for your datacenter needs. So here is a few links to help begin the discussion.
I am avoiding a post where I have to think really hard about a topic. That makes me procrastinate and come up with even crazier ideas. I am writing this one down now. Most of these apply to me so if you are offended by any of them you are probably a vDiva.
You might be a vDiva if…
… you roll your eyes when someone talks about installing a PHYSICAL server.
… you are on twitter to see how many people you can get to look at your blog, but you never stoop so low to interact with the common folk.
… you are surprised when the guy at the table at the VMUG doesn’t know who you are.
… you constantly check your Google Analytics account to see how many views you have. (guilty)
… you refer to yourself as @… (your twitter account)
… you hunt down @jtroyer if you latest post takes too long to get on the v12n board.
… your require a signed rider agreement with your speaking topic for VMworld, saying you need 800 green M&M’s, a copy of Lord of the Rings in your hotel room, and direct phone access to Steve Herrod’s iPhone.
I probably ticked a bunch of people off. I am just having fun. Have a great day! Go ahead and add your own in the comments.
I try to not “self promote” too much. A co-worker and I submitted a topic in the Desktop Virtualization track and I am giving in and spreading the word:
Thinning Down to Scale Out
Abstract:
Desktop Virtualization provides the ability run hundreds even thousands of desktops. Each small performance enhancement can make a difference when multiplied across an entire enterprise. This presentation will demonstrate the steps necessary to thin down your guest desktop image in order to provide overall better user experience.
Speaker: Kevin Miller, VeriStor Systems
We are not popular so this doesn’t qualify for the popularity contest that many other sessions have, plus my name doesn’t even show on the Session. One thing I will say is this will be a Practical technical session. You should leave saying, “here is some stuff I can do to improve my VDI setup.”
So go vote and if we get to present the session I will give you a high five. Vote now because voting ends May 26.
One thing I am thinking about due to the VCDX application is operational readiness. What does it mean to pronounce this project or solution good-to-go? In my world it would be to test that each feature does exactly what it should be doing. Most commonly this will be failover testing, but could reach into any feature or be as big as DR plan that involves much more than the technical parts doing what they should. Some things I think need to be checked:
Resources
Are the CPU, Memory, Network and Storage doing what they should be? Some load generating programs like IOmeter can be fine to test network and storage performance. CPU busy programs can verify Resource Pools and DRS are behaving the way they should.
Failover
You have redundant links right? Start pulling cables. Do the links failover for Virtual Machines, Service Console, and iSCSI? How about the redundancy of the physical network, even more cable to pull! Also test that the storage controllers failover correctly. Also, I will make sure HA does what it is supposed to, instantly power off a host and make sure some test virtual machines start up somewhere else on the cluster.
Virtual Center Operations
Deploy new virtual machines, host and storage VMotion, deploy from a template, and clone a vm are all things we need to make sure are working. If this is a big enough deployment make sure the customer can use the deployment appliance if you are making use of one. Make sure the alarms send traps and emails too.
Storage Operations
Create new luns, test replication, test storage alarms and make sure the customer understands thin provisioning if it is in use. Make sure you are getting IO as designed from the Storage side. Making use of the SAN tools to be sure the storage is doing what it should.
Applications
You can verify that each application is working as intended within the virtual environment.
There must be something I am missing but the point is trying to test out everything so you can tell that this virtualization solution is ready to be used.
