For a great overview of the entire U1 release check out Rick Vanover’s post vSphere 4.1 update 1 released from his Rickatron Blog and via his Servers and Storage Column/Blog at TechRepublic.

The rest of this post contains cut and pastes from the Release Notes and some commentary about them from me. I want to stress again that these are issues that have now been fixed!

Finally, I’ll point out the one huge VM backup issue (that I can think of right now) that still does not appear to be resolved.

• Cannot take quiesced snapshots of Microsoft Windows Server 2008 R2 virtual machine running vCenter Server 4.1 When creating a snapshot of a Microsoft Windows Server 2008 R2 virtual machine that has vCenter Server 4.1 installed, the snapshot operation might fail to complete. This issue occurs on Microsoft Windows Server 2008 R2 virtual machines when the ADAM database is installed. The issue is resolved in this release.

The change from 2008 to 2008 R2 has caused quite a few applications to stumble, so it is good to see that VMware users can now install vCenter on the latest and greatest OS from Microsoft running as a virtual machine without having to worry about various hassles, not to mention coming up with a separate backup or business continuity solution for arguably one of the most important VMs in the environment.

• Creation of large .vmdk files on NFS might fail. When you create a virtual disk (.vmdk file) with a large size, for example, more than 1TB, on NFS storage, the creation process might fail with an error: A general system error occurred: Failed to create disk: Error creating disk. This issue occurs when the NFS client does not wait for sufficient time for the NFS storage array to initialize the virtual disk after the RPC parameter of the NFS client times out. By default the timeout value is 10 seconds. This fix provides the configuration option to tune the RPC timeout parameter using the esxcfg-advcfg -s <Timeout> /NFS/SetAttrRPCTimeout command.

Although specific to users of NFS storage, this one was definitely a head scratcher. Possible scenarios where this problem could rear it’s ugly head include full VM restores and even VM replication. Add the possibility of dedicating a large .vmdk as the backup repository for your backup server VM.

Let me know if I missed any other resolved issues specific to VM backups.

The issue still unresolved

Unfortunately, nothing in vSphere 4.1 U1 appears to address the loss of connectivity when committing snapshots for VMs using Change Block Tracking (CBT) while running on NFS storage.  This leaves VMware admins stuck between a rock and hard place – disable CBT or move VMs off of NFS. More about the problem can be found in VMware’s KB Article:

Virtual machine freezes temporarily during snapshot removal on an NFS datastore in a ESX/ESXi 4.1 host

© Rich for Gestalt IT, 2011. | vSphere 4.1 U1 Released. Fixes Specific For VM Backups
Read more posts categorized as All, Server Virtualization

YouTube Guinness Brilliant Six Pack Commercial

It’s obvious now that Cloud Computing is no longer just a concept. Amazon EC2 has been around for a while, vCloud Express was announced late last year, Microsoft is moving full steam ahead with Azure, and new internal cloud infrastructure and storage solutions are appearing on the scene weekly. CTOs and IT Directors are starting to see legitimate solutions for offloading some or all of their development and production workload and infrastructure to alternatives in The Cloud.

Put yourself at the helm of one of the Original Equipment Manufacturer’s (OEMs) that have made their money selling server, switch and storage hardware in the private data center to date. Faced with the future possibility that companies will have an option to run applications and services on infrastructure they don’t buy, build, or maintain, what would be your strategy for generating reoccuring business in the future?

I bring it up because of all the cloud architecture announcements. EMC announced a coalition with Cisco and VMware, NetApp has a Secure Multi Tennancy alliance with Cisco and VMware, and HP has announced an Integrated Infrastructure partnership with Microsoft. The storage OEMs are the first out the gate with the snap together infrastrucure for the cloud, but I imagine other hardware partnerships are not too far behind. VMware and Cisco UCS may already be the first with Microsoft and HP? Microsoft and VMware conveniently can run on top if it all.

Before the Alliance, Coalition, and Partnership start the Cloud Wars, everybody raise your Guinness and say “Brilliant!”

It’s brilliant because it’s a simple win / win solution no matter what happens. The appeal of Everything as a Service, Iaas, Saas, Haas, and any other “as a service” you want to think up is interesting and efficient whether implemented internally or externally. Those hardware and software providers that have a cookie cutter architecture with guidelines for implementing in a dynamic, do it yourself, modular data center should stand out as go to manufacturers.

The point is that, although the OEM partnerships may have many complex implications that are being discussed in the community from every possible angle, at the heart of the announcements is a simple desire to keep selling products. Selling them no matter who is buying – the traditional IT department or the cloud provider. Like the Guinness six pack, combining the equipment also results in more unit sold for everybody.

Brilliant!

© Rich for Gestalt IT, 2010. | If You Were An OEM Facing The Cloud What Would You Do?
Read more posts categorized as Cloud Computing, Server Virtualization

I explained my most recent personal evolution of media gadgets to point out that many devices are podcast capable, so even if you don’t have an iPod you probably can still subscribe with most any portable device. The point is any mp3 player could be used to broaden your virtualization and general technical knowledge. The trick is figuring out how to find, subscribe, and sync the content to your device with the least amount of effort.

For those looking for other software options,

I’ll quickly mention a couple of alternatives to iTunes with native ability for subscribing to podcasts or transferring files to your device:

• Winamp
• MediaMonkey
• doubleTwist
• CopyTrans Manager

Winamp requires a plug-in to work with iPod Nanos and earlier models, but my recently updated iPod Touch 3G doesn’t seem to work yet. I never tested the Touch 2G and Winamp personally. MediaMonkey and doubleTwist read my iPod Touch just fine. All 3 can sync with my iPods, and Blackberrys, and according to their web sites just about all other mp3 players and cell phones with media playing features.

Since the other options do not seem to handle video as well, I primarily use CopyTrans for manually transferring mp4s. CopyTrans does not allow you to subscribe to podcasts, however.

In some cases when your media player software is not able to do the job, you can use a RSS reader to subscribe and download podcasts and then sync to the player as you would any other mp3s.

I thought I would list the Virtualization, storage, and technology podcasts that are currently in my sync schedule and hopefully get some feedback and recommendations from VM /ETC readers – both on great podcasts I’m missing and other tips for transferring and listening to them. This is by no means an exhaustive list. Let me know about other great podcasts I might be missing!

Maybe I can convince Eric Siebert to add a new category to his popular vSphere-land.com Top 10 Lists and rankings?

My Current Podcasts (in alphabetical order)

I’m providing both a link to each podcast web site as well as the RSS feed. I’m also including the description provided by the podcast hosts themselves. Follow the site link to find the options for subscribing with your media player software or enter the RSS feed to start downloading the podcast .mp3s to your computer via your favorite feed reader.

• BrianMadden.com Daily Podcast

This daily podcast is like an audio version of BrianMadden.com. In each episode we discuss the day’s main story and the conversation from the previous day.

http://feeds.feedburner.com/brianmadden/rss

• Cranky Geeks

CrankyGeeks, hosted by PC Magazine contributing editor John C. Dvorak.  Each week a group of expert panelists discuss the most controversial topics in tech. CrankyGeeks airs on Wednesdays at 3:30pm eastern, and can be downloaded at www.CrankyGeeks.com

http://www.crankygeeks.com/index.xml

• gdgt weekly

You may know Peter Rojas and Ryan Block as the guys behind Engadget and Gizmodo, but they’re back with gdgt – the ultimate online destination for all things gadgety and electronic. Tune in every week as Peter and Ryan discuss the latest goings on in the world of gadgets.

http://feeds.gdgt.com/gdgt/podcast-mp3/

• GeekBrief.TV

Geek Brief TV is a Video Podcast with news about tech tools and toys featuring Cali Lewis. The format is fast, fun, fresh and flirty.

http://www.geekbrief.tv/feed/

• Infosmack

Greg Knieriemen and Marc Farley discuss current events and strategies around the data center with the leaders in the industry. Topics include data storage, networking, storage virtualization, server virtualization (VMware) and backup.

http://infosmack.podbean.com/feed/

• Storage Soup

A SearchStorage.com podcast covering the top stories in enterprise data storage from week to week, also featuring interviews with industry experts.

Storage Soup Podcast RSS Feed

• TechStuff Podcast

How do fiber optics work? How about the Sony PSP? Join Chris and Jonathan as they discuss a wide array of technological topics and trends in TechStuff, a podcast from HowStuffWorks.com.

http://www.howstuffworks.com/podcasts/techstuff.rss

• Tekzilla

Tekzilla feeds your tech hunger! Patrick Norton and Veronica Belmont bring you hands on reviews of the latest gear, tips and tricks that improve the tech you already own and conversations with the most informative experts around.

http://revision3.com/tekzilla/feed/quicktime-large/

• This Week In Virtualization

Get virtualization news and find out how it affects IT managers and systems administrators in this weekly server virtualization podcast. Learn about the latest server virtualization market news from VMware, Microsoft, Citrix Systems and other virtualization vendors.

http://itknowledgeexchange.techtarget.com/this-week-in-virtualization/?feed=podcast

• VMware Communities Roundtable

VMware experts talk about VMware news, interesting virtualization posts, and what’s happening in the VMware Communities.

http://recordings.talkshoe.com/rss19367.xml

I recently found the post 15+ Podcasts Every Geek Should Listen To. The list contains several of the podcast I’ve already mentioned, but several more that I have not checked out myself yet. I plan on adding a few to my subscription list and see what sticks!

As I mentioned previously, I want this post to not only be my podcast suggestions to readers but I am searching for new content myself. Let me know your suggestions!

© Rich for Gestalt IT, 2010. | Options For Virtualization, Storage, And Technology Podcasts
Read more posts categorized as Server Virtualization, Storage

All that being said, I wanted to offer my thoughts on what I think could help both sides during future GestaltIT Tech Field Days (yes, it looks like there will be others!)

Watching CLI Demos Is Like Listening to the Ballgame On The Radio

I’m not saying using CLI demos, configuration, or hands on labs are bad, but watching text scroll down a screen is not exactly engaging either. If CLI commands are the method of configuring the product’s features then so be it, but

Rather, after you tell me the product has a capability then show me the results from the user’s perspective, or in other words, a screen shot of “My Computer” or “Disk Manager” before and after better explains what actually happened. That’s a simple example. But the point is a few images are worth a thousand commands when you want to high-lite what the product does.

The Server and Network Guys Need To Understand What It Needs Before They Care About How It Does It

Look, I know the GestaltIT crowd is mostly storage gurus, and as one of a few server and network engineers I’m not who the sponsors prepared for. But, don’t count me out! Fact is, your storage product needs servers and networking to do it’s thing. My kind needs to understand the infrastructure your product requires before we entertain your technology.

Many of the company’s products I learned about this week were first time introductions for me. By nature of the GestaltIT Field Day’s objective, the sponsors of the future events will be lesser known technologies too. Take some time to explain a logical view of your solution. I’m not talking about the marketing either (although GestaltIT attendees have the expertise represented too). For example, if you have a hardware appliance that optimizes my storage than show me a Visio, photo, or whiteboard where to put it in the data center. Be sure we understand the implementation and connectivity first. Then I’m ready for a deep dive on why your technology works, but until I can visualize where it sits in the rack and what talks to it, I’m content to take your word that your “black magic” works – and yes, telling me it is “black magic” is OK while I’m understanding where it goes. Show me the actual “ingredients” of your “potion” at the end. In some scenarios, the tour of the data center and the hands on exploration of the hardware could be more effective before the presentation as opposed to the traditional “on the way out” time slot.

My 2 cents: a suggested outline for presenting

• Welcome – Introduction to product
  • use case: how technology solves specific problems
  • target customers
  • Infrastructure architecture and requirements
• Photos, samples, tour, etc. (get the audience out of their seats!)
  • Model nos. and their capabilities
• Technology deep dive (sit them back down)
• Hands on Labs, demos, etc (have support for questions beyond the scope of the script)

We’re going to debate about using the technology, and that’s a good thing

In the end, the best part about the group of Tech Field Day attendees is that we are a mix of technology perspectives, and those perspectives created invaluable discussions on the bus, at the dinner table, and at the bar. Those talks are where it all comes together for us. The group discussion was so beneficial that we are considering building in time between sessions in future event scheduling to allow us to meet and debrief.

To illustrate the importance of  this, there were several sessions this week where I personally missed the point. I either tuned out, was distracted with email or tweets, or I couldn’t visualize the way the technology made a difference. However, later each day I got the chance to question those in the group that understood – sometimes to the frustration of many. As a collective we struggled to understand what we saw from the holistic perspective: networking, storage, and servers. The result was that everyone ended up considering the technology in ways slightly differently than our own, first impressions. The key is the chemistry of the group allows these debates!

Help us have these discussions by making sure all the data center perspectives (server, networking, security, and storage) are covered by the presentation.

© Rich for Gestalt IT, 2009. | Tech Field Day: Thoughts About Presenting To Engineers
Read more posts categorized as Events, Server Virtualization

Confused? Me too. Time will clear the fog and, as promised, reveal the vBlock based Cloud.

VCE vBlock is big, bad, and designed for scale like VMworld 2009 Infrastructure (without being told as much, my bet is that what we saw at the bottom of the stairs in the Moscone Center was a vBlock test drive). Apparently, VMworld’s spotlight on vCloud Express provider Terremark was another hidden VCE vBlock showcase from Vmworld. It all was happening in plain sight. If we only knew then what we know now. 

My thoughts (or this rant) can best be described as “now what!?”. I’ve spent the last 5 years figuring out how to design customized, application specific, performance optimized, and product feature specific virtual infrastructure designs for the enterprise data center. This is because I’ve listened to everybody explain that a “cookie cutter” infrastructure is not adequate for their solutions. Change this setting for feature A, provision these LUNs for feature B, add these VLANs for feature C, etc., etc. Now, VMware, Cisco, and EMC have figured out all the tough stuff and come up with a “connect the dots” data center reference architecture to make it easy to move to the Cloud? Really? I mean, I know we’ve been told this had to happen in order for the Cloud to work, but REALLY?!! Oh well, the only constant is change.

Ok, I feel a little better. Anybody else care to vent?

Here’s some more random links and quoted information to help others with similar change anxiety issues as me.

From the Cisco press release: Cisco and EMC, Together with VMware, Form Coalition to Accelerate Pervasive Virtualization and Private Cloud Infrastructures The following family of Vblock Infrastructure Packages is being offered by the Virtual Computing Environment coalition:

• Vblock 2 is a high-end configuration supporting up to 3,000-6,000 virtual machines that is completely extensible to meet the most demanding IT needs of large enterprises and service providers. Designed for large-scale and ‘green field’ virtualization, Vblock 2 takes advantage of Cisco’s Unified Computing System (UCS), Nexus 1000v and Multilayer Directional Switches (MDS), EMC’s Symmetrix V-Max storage (secured by RSA), and the VMware vSphere platform.
• Vblock 1 is a mid-sized configuration supporting 800 up to 3,000 virtual machines to deliver a broad range of IT capabilities to organizations of all sizes. Designed for consolidation and optimization initiatives, Vblock 1 is comprised of a repeatable model leveraging Cisco’s UCS, Nexus 1000v and MDS, EMC’s CLARiiON storage (secured by RSA), and the VMware vSphere platform.
• Vblock 0 will be an entry-level configuration available in 2010, supporting 300 up to 800 virtual machines, for the first time bringing the benefits of private clouds within reach of medium-sized businesses, small data centers or organizations, and for test and development by channel partners, systems integrators, service providers, ISVs, and customers. Vblock 0 is also comprised of a repeatable model leveraging Cisco’s UCSand Nexus 1000v, EMC’s Unified Storage (secured by RSA), and the VMware vSphere platform.

Cisco VCE vBlock Youtube video

The Coalition leaders calls VCE vBlock “IT as a Service”

Chuck’s Blog: Announcing The VCE Coalition

Introducing The Vblock

Building virtualized infrastructure at scale?  We’ve created a new option for you — the Vblock.

We’ve taken the best of the best from the three respective companies, and create a pre-architected and pre-qualified environment for virtualization at scale: storage, fabric, compute, hypervisor, management and security.  We’ve also characterized its behavior for specific use cases as well.

Whether you actually want to buy one, or just use the reference architecture as a blueprint for your own efforts — there’s an entirely new option available to accelerate the transition to a fully virtualized environments.

© Rich for Gestalt IT, 2009. | Does VCE vBlock Really Mean Cookie Cutter Architecture For The Cloud?
Read more posts categorized as Cloud Computing, Server Virtualization

I am using terminology and acronyms established in my post from day 1. Review that post if necessary.

Upgrading Firmware

For those familiar with management of bladecenter chassis, blades, and modules from the leading manufacturers today UCS hardware is a bit of change. Specifically, the “management module” is not found in the chassis. Management is instead performed on the Interconnect switches via the UCSM GUI or CLI. Administrators used to configuration, troubleshooting, and upgrading hardware at the chassis level could be surprised to learn that even though there is a Chassis Management Console (CMC), it is not directly accessible without a special connector possessed only by TAC or authorized Cisco Partners. UCSM controls the CMC and orders it to make the configuration changes.

Once configured, a web browser is pointed at the switch ip address and all components are upgraded from there.

Firmware upgrades can be performed via the CLI, but using the UCSM GUI makes the entire process, from downloading the update bundle to applying individual component upgrades, much easier.

All of the following can be upgraded from the UCSM:

• Mezzanine Cards (IOMs)
• Blade BMC
• Blade BIOS
• Chassis CMC
• Interconnect switches
• Blade LSI RAID controller

The upgrade bundle download is initiated from the UCSM and then is viewable in the interface, and the installable images can be assigned to the hardware.

Deploying Server Profiles

As established on day 1, the Server Profile is the “soul” of the UCS Blades, and today we were introduced to the concept of provisioning blades like traditional servers (without stateless mobility). Cisco calls this “the basic Opt-In Model”. The lab exercise outlined at the end of this post illustrates a ” Opt-In Model” install. The point to note, regardless of stateless install or not, is that without a Server Profile the host will boot but it will not have LAN or SAN connectivity. Therefore profiles are needed in all scenarios. Understanding the Server Profile is the key to UCS blade management. Here’s a quick review of some key points on UCS Server Profiles:

• Server Profile represents s a logical server and stores personality and connectivity requirements
• A Server Profile is associated with one blade at a time
• Profiles can be cloned or created from template

UCSM is used to create the profile and establish the WWN, MAC, UUID, NICs, HBAs, bios boot order, etc. UCSM lets you assign the created profile to a UCS blade. Looking under the hood at how UCS actually deploys a Server Profile reveals a semi-complex system of VLANs, PXE boot, and the PnuOS. We were not specifically told that any configuration was required to create this profile distribution system so I’m assuming the following pieces described in the process are ready “out of the box” on the Interconnect switches.

• UCSM powers on a blade if it is not already
• UCSM reboots the blade (I assume some control change occurs at this point but I do not recall exactly what)
• UCSM provides a PXE boot of the blade to the PnuOS on VLAN 4047- PnuOS is a specialized Linux OS provided/created by Cisco
• UCSM communicates to the PnusOS to make the changes
• UCSM configures UCS fabric for VLAN and SAN

The Part of Tens – Screen shots

Deploying an “Opt-In” Profile and installing ESX 4 on UCS Blades

Note: this is not a “how to”. It is a list of highlites for the process. These screen shots are of UCSM in Firefox web browser running on Windows 7. Java 6 is required.

UCS Blade not associated with a Server Profile

Creating a Server Profile

Associate blade with Server Profile

Blade PXE boots to PnuOS

UCS Utility OS Boot Complete

Mount ESX ISO from client connected to UCSM

(blade already running W2K3)

ESX 4 install boots on UCS Blade

CNA shows up as 10 GE NIC in Network install screen

ESX 4 install complete

© Rich for Gestalt IT, 2009. | Cisco UCS for Dummies – Managing Blades With UCS Manager
Read more posts categorized as Server Virtualization

During the UCS Bootcamp in San Jose, Cisco made it clear that the value proposition of UCS is the Stateless Model. Unlike traditional server deployment use of the Service Profile (I covered the Opt-In Model earlier in this series), the Stateless Model allows the physical hardware to become generic and, since the operating system and application resides on the SAN, a server personality duplicated and restarted from blade to blade.

Cisco’s definition of Statelessness:

“In Unified Computing system (UCS) the underlying hardware (or server) can be made completely transparent to the OS or applications that run over it. The kind of environment which an OS or application requires can be moved from one server to another or can be changed very easily. This is made possible by moving resources, such as MAC addresses, WWN values, IP addresses, UUID, firmware versions and even server BIOS, from one server to another at the time of deploying the server. This is accomplished by using the concept of Service profiles; which is like software definition of a server. The concept of stateless computing facilitates much greater scalability and can be used in conjunction with virtualization to achieve maximum data center utilization.”

One of the labs during the week showcased the Stateless Model in action, so what better way to help explain this feature then to walk through it again for all to understand?

The Stateless Model Lab Overview

Quoting the lab introduction, the purpose of the lab was to:

“.. demonstrate the statelessness by booting an OS off of a SAN LUN.  The SAN connectivity and masking is specified by World Wide Names that are associated with the service profile.  When your service profile moves from one blade to the next, you will be booting the exact same SAN based OS. No configuration outside of UCS will ever be required at this time.”

The following overview is of the UCSM configurations performed in the lab. Once again, this is not a “how to” but is instead intended to provide insight into the process and advantages of the UCS Stateless Model.

Create a Server Pool

Select multiple UCS blades to be in a pool. Almost is if the hardware was like non persistent virtual desktops and UCSM was the user, Service Profiles will be able to move between the hardware in a pool allowing OS and application to run on any pool member without any further setup. Cisco referred to these pools as “server farms.”

Create WWPN and WWNN Pools for storage

SAN ports and initiators are also to be grouped as pools. When server profiles move to another blade your FC fabric and storage see no change. No remapping will be required.

Cisco defined the SAN boot components configured in these pools the “triplet” because it includes:

• vHBA name
• WW port name of target array
• LUN#

For multipathing 2 triplets can be specified in the boot order. Remember that the mezzanine cards do not provide multipathing, but the operating system is instead responsible.

Create a MAC Pool for networking

Network interfaces are pooled as well. The MAC pool will be the vNICs used by the UCS blades in the pool. vNICs already have a native VLAN and allowed VLANs assigned, so  the networking configuration remains seemless and mobile with the blades. This is matched to the configuration on the Northbound switches already created by the network administrator.

Create Service Profiles to use Pools

The steps from the lab explain the use of the pools created.

1. Create service profile (explained with screen shots in Managing Blades with UCSM post)
2. For the World Wide Node Name, choose the WWNN name pool.
3. For the vHBA, choose a WWN from the WWPN pool
4. For the vNIC, choose a MAC address
5. Set the boot order.
6. Associate your service profile by choosing the blade server pool, rather than a specific blade.

The creation of the Service Profile and the association of a blade causes the PXE boot of the PnuOS where the profile is applied, and then the blade can access the assigned LUN(s). Obviously, the SAN administrator has pre created LUNs and configured appropriate masking/zoning in order for this to work properly.

Clone the Service Profile

In the UCSM the service profile can be right-clicked where the administrator can choose to create a clone. Once again, the LUNs have been pre masked/zoned, and since the cloning process assigns a new value from the WWN, MAC, and server pool the cloned service profile results in an independent blade operating system and workload.

Summary

Matching the assigned configurations from the storage and network administrators is crucial, but once in place the UCS (server) admins handle all inbound connectivity set up. Thus, hardware mobility is enabled through pools.

Pooling servers that will be dedicated to similar functions, like ESX hosts for example, allows for workload mobility across reserved hardware. Cisco called this hardware availability, and stressed this is not the same concept as high availability. If you recall from the previous posts that moving server workloads is a manual process that requires OS shutdown, then it makes sense that virtualization is needed for the true high availability scenarios.

In a non virtualization example, consider a database run on a UCS blade pool. Since there are reserved blades in the pool proactive monitoring could identify a hardware failure developing, and the database could be powered off and relocated to reserve blade hardware. Although many IT shops do this today with bladecenters, the advantage in UCS is that the SAN, Network, and even the OS and application changes do not need to be performed, and the new hardware is then more quickly running the workload.

Just like the option to clone a service profile, a template can also be created. Templates come in two flavors, Initial templates and Updating templates. Initial templates seemed like the more common usage scenario similar to deploying new virtual machines in vCenter for example. Updating templates are a bit complex, remain linked to any Service Profiles created from them, and therefore changes to Updating Profiles impact blades in service. We did not cover these in detail in class so I am still a little fuzzy on templates.

I’ve got one more post for this series in mind. I plan to address the questions I outlined in my first UCS for Dummies post and explain some of my final thoughts on how, where, and why UCS fits in the datacenter.

© Rich for Gestalt IT, 2009. | Cisco UCS for Dummies – The Stateless Model
Read more posts categorized as Server Virtualization

Again I am using terminology and acronyms established in my post from day 1. Review that post if necessary.

The following diagram illustrates the current connectivity between the UCS Blades, Fabric Extenders (FEX), and the Interconnects. The diagram only includes a single chassis, a single half height blade, and a single full height blade for simplicity while covering all scenarios. Duplicate the same connectivity for each blade inside the chassis, and duplicate connectivity of 2 more FEX for each additional chassis in the solution. As shown, the 2 Interconnects can manage up to 20 different chassis if model 6140 and up to 10 chassis if model 6120. (The max number of chassis can not be achieved because 2 FCoE cables are being used to the Interconnects)

Click on the image to see a larger version.

New terms to understand before continuing:

• Northbound networking – any connectivity and communication to other switches outside of the UCS solution. Port channels, allowed VLANs, and a matching native VLAN must exist on the next switch up. LACP will be automatically configured. vSAN object in Service Profile
• Inbound networking – any connectivity and communications to blade servers. Configured by the UCSM, assigned via Service Profiles and represented as a vNIC or vHBA objects. Includes MACs, native VLAN, allowed VLANs, WWNs, WWPNs, etc.
• End-host mode - UCS Interconnect default operation. No MAC tables are maintained. Does not switch traffic northbound, but does switch traffic inbound – including both blade to blade connectivity and packets from outside UCS headed inbound.
• Pinning – automatic or manual assignment of ports. Happens both on the Interconnects and the mezzanine cards. On the interconnects  without switching or MAC tables for northbound traffic from blades.

Yes, other switches for the LAN and SAN are needed since the Interconnects do not route or switch, and FCoE adapters in the storage device cannot be directly connected to the 6100s.

There is an option to change the mode of the entire UCS switch to “switching mode”, but it is highly recommended not to do this.

Blades cannot communicate with each other inside the same chassis via the FEX. Local traffic must travel to the Interconnects first.

There is no multipathing provided from the blade hardware (mezzanine cards). Multipathing is only possible from the blade operating system.

On the Interconnects, only ports 1 through 8 are licensed by default. ports 9 through 20/40 are licensed per port as needed.

Oplin mezzanine cards provide ethernet only. Menlo and Palo provide both LAN and SAN connectivity.

FCoE Boundary

The biggest misconception I’ve had about UCS (and it has been common among a lot of people I have talked with) is where FCoE is used in the solution. In the current version of UCS FCoE exists between the mezzanine cards on the blades to the Interconnects only. FCoE is not possible between the Interconnects and the northbound switches. As mentioned earlier, a FCoE adapter in a storage device cannot be directly connected to the Interconnects. This is possibly on the roadmap, but today’s UCS cannot do it.

Blade mezzanine card to FEX connectivity

Each mezzanine card has 2 ports both capable of 10 GE. Half height blades can hold one mezzanine card and full height blades can hold two – or 4 ports each capable of 10 GE.

Without 2 FEX in a chassis only one mezzanine port will be active per card. This means fail over is not possible for half height blades and only possible in full height blades if two mezzanine cards exist – 1 active port on each card.

Without 2 interconnects having 2 FEX is useless. You can not connect both FEX to the same 6100.

Pinning also occurs between the mezzanine card and the internal ports on the FEX (inside the chassis). This assignment of ports is automatic and depends on the number of cables between the FEX and the Interconnects. Only 1, 2, or 4 cables (ports) can be used and pinned as follows:

• 4 cables
  • Blade 1 and 5 to port 1
  • Blade 2 and 6 to port 2
  • Blade 3 and 7 to port 3
  • Blade 4 and 8 to port 4
• 2 cables
  • even numbered blades to port 1
  • odd numbered blades to port 2
• 1 cable
  • all blades to single port

If you have 4 cables uplinked and one fails UCS will have to re-pin the blades to a 2 cable configuration. Blades using ports 3 and 4 will temporarily lose connectivity.

The Part of Tens – UCS was built for virtualization

Blades cannot communicate inside the chassis – If the 10 GE between the FEX and the interconnects is not enough bandwidth for an application, running ESX on the blades allows an affinity to keep VMs that need local connectivity together.

Operating system multipathing only – VMware vSphere to the rescue again.

Hardware high availability limitations – vSphere VMotion, DRS, and HA server this purpose.

Bandwidth reduction from the interconnects to the northbound switches – Virtualized servers, regardless of chassis location, managed by the same interconnect domain should rarely have northbound needs. Until physical clients have 10 GE adapters inbound network traffic will not be an issue. Some storage devices currently have FCoE adapters however, and Cisco is aware of the need but maintains current virtual server loads do not need that size pipe.

© Rich for Gestalt IT, 2009. | Cisco UCS for Dummies – LAN and SAN Connectivity
Read more posts categorized as Server Virtualization

I promised to report in the style of the “For Dummies” series. To do that I am going to borrow the common feature “The Part of Tens” always found at the end of those books. The “Tens” are usually a list of concepts suggested for further research. I’m going to use this more as a list of what “sticks” with me from the bootcamp topics. As in the books I hope I motivate VM /ETC readers to research further. If you count My “Tens” it might not even equal 10 items – no promises. I’ll add some opinion and scenario when motivated to do so.

Many others already have provided technical details on the Cisco chassis, blades, modules, adpaters, and the other components, and I’m not about to add anything new to what is already known. In fact, I found myself supplementing what I was being told in class by cross checking various blogs ande other documntation already on the web. Nevertheless, the first “Tens” covers the overview of the UCS parts and concepts.

Cisco UCS Overview Part of Tens

• Converged Network Fabric - simply put, this means combining the ethernet and fiber channel (FC) connections from the blade chassis over the same cables. To do this Cisco’s Nexus switching uses 10 gigabit ethernet (GE) to encapsulate FC frames within ethernet frames. This is called Fiber Channel Over Ethernet (FCoE). FCoE effectively can cut the amount of cables needed in half.
• UCS Blades – have 2 cpu sockets, 2 internal SAS drive bays, slots for mezzanine adapters, and DDR3 DIMM slots.  Half width blades have 12 DIMMs and full blades have 48 DIMMs for up to 384 GB of RAM per blade.
• Converged Network Adapters (CNAs) – Cisco’s mezzanine adapter cards that are inserted in each UCS blade in order to connect to the Converged Network Fabric. Cisco currently has 3 adapters, Oplin, Menlo, and Palo, which appear to the blade operating system as 2 GE NICs and 2 4GB HBAs. UCS half blades can hold 1 mezzanine card and full blades can hold 2 mezzanine cards.
• UCS Blade chassis – 6U tall and 32″ deep. Can hold 8 half width blades or 4 full width blades or a combination of both.
• Fabric Extenders – The UCS chassis holds the  Fabric Extenders (FEX) otherwise known as IO Modules (IOM) that connect the blades to the switch. There are 4 10Gbps ports per FEX, and 1,2, or 4 connections are allowed to the Fabric Interconnects.
• Fabric Interconnects – switches the fabric extenders from the chassis uplink to. Can be 20 port or 40 port. Interconnects have connectivity to both the LAN and the SAN networks.
• The Full UCS System – consists of 2 Fabric Interconnects, 1 – 40 chassis with 2 FEX each, and 1 – 8 blades per chassis.
• UCS Blade Statelessness – based on creating Server Profiles defined as what Cisco calls “the behavioral and identity elements that make the server unique to all others.” Cisco listed WWN, MAC, and UUID as identity and boot order, firmware versions, and QoS as behavior. The server profiles include the operating system and can be applied to bare metal UCS blades for fail over scenarios. UCS blades are required to boot from SAN to enable profiles, and the process is manual only. Blades must be powered off before they can fail over. Cisco does provide APIs for third parties to develop automation.
• UCS Manager (UCSM) – runs on the Fabric Interconnects and accessible via a web browser as a Java application. It is a single pane of glass to manage the entire UCS system including the switches, the chassis, and the servers. Via roles and pemissions administrators can be assigned specific access. For example, the team that manages the interconnect switches does not need to access the chassis and servers and vice-versa.

Final Day 1 UCS Thoughts:

Overall I was impressed with the UCS and the ability to accomplish from the UCSM what requires multiple management interfaces and applications in other blade hardware. Stateless Server Profiles seem to offer simplified migration of server personality between blades. I witnessed first hand the power of the UCS hardware and the significant reduction of cables at VMworld.

One of my classmates made a great point today about comparing UCS to the other bladecenter offerings – UCS’ advantages are greatest in a large grid deployment. One or two chassis in a rack do not seem to be a good scenario from a competitive perspective. I’m not making final judgement on this statement just yet as there may be specific features not revealed yet that may justify UCS for smaller environments.

FCoE is the obvious push. I’m sure there will be more about this later in the week.

UCS hardware was made for VMware virtualization with the ability of each blade to provide large amounts of RAM and CPUs for any application.

Posts that diagram the full UCS solution:

http://rodos.haywood.org/2009/08/ucs-schematic-sketch.html

http://www.internetworkexpert.org/2009/07/05/cisco-ucs-vmware-vswitch-design-cisco-10ge-virtual-adapter/

© Rich for Gestalt IT, 2009. | Cisco UCS for Dummies – UCS Overview
Read more posts categorized as Server Virtualization

VDI has gained momentum. VMware, Citrix, and Microsoft all boast about the availability of an enterprise ready virtual desktop management solution. Storage technologies such as thin provisioning, deduplication, and rapid cloning are being perfected and optimized and will lower the total cost of ownership for VDI. 64 bit Hardware with virtualization assist technologies on host servers has enabled record setting ratios of supported users. It seems that OS licensing models formerly based on physical hardware are slowly morphing to support virtualized environments. But, in my opinion, there is still a major hurdle to the enterprise’s wide spread acceptance of VDI on a platform other than Windows.

Is enterprise business really ready to migrate the majority of their user applications and data to Linux, the web, or even the Cloud? For Chrome OS to challenge Microsoft for seats in the enterprise, this shift will have to happen, and it will have to be publicly reported as highly successfully and common. Are web based applications and software as a service (Saas) alternatives ready to handle main stream business services like email and office collaboration? More importantly, are enterprise IT departments ready to allow cloud providers to run, protect, and store their business critical applications and data? When that happens the OS on the desktop becomes less important. The browser will become the only application of importance.

Until then, Windows will continue to be the most popular desktop OS simply because it runs the most popular applications needed by business today. Therefore, the typical VDI solution in the immediate future will continue to consist of Windows VMs. Microsoft’s decision to strip down the bloat and replace Vista with Windows 7 was a wise move, and probably has ensured Microsoft’s position as the desktop market leader for many more years to come.

That’s why it more important to think about what will the users use to connect to those virtual desktops. Yes, VMware, and the competition, is developing their own desktop hypervisor OS. VMware calls their work in progress CVP (Client Virtualization Platform), but is VMware’s CVP (for example) really going to be able to run on all modern notebooks and desktops? To be specific, on the Lenovo T400 I am using to write this, will a VMware client hypervisor allow all of the volume buttons, wifi adapter controls, microphone, blue tooth, and other miscellaneous hardware features to work properly? What about all the other manufacturers of notebooks, netbooks, and desktops with their different models and hardware configurations too? This is where all the work put in to a modern Linux distribution appears to have the advantage. Otherwise, does the future client hypervisors requiring their own specialized hardware really sound appealing? Does expecting users to use today’s notebooks without the full functionality sound realistic?

Assuming Google’s product will be most like a Linux distribution, it seems to me that Google’s best bet for mass adoption of the Chrome OS would be to make sure that enterprises can count on installing VMware’s View Client (and the competition’s VDI clients as well). The real corporate desktop will still have to be a Windows virtual machine that can be synced between the data center and the client OS and even used offline. Eventually, as cloud based application replacements emerge in the future, Google’s OS could become more of a direct competitor/replacement to Microsoft’s OS.

© Rich for Gestalt IT, 2009. | VMware Will Virtualize Google Chrome, But Will Chrome Run VMware For VDI?
Read more posts categorized as Desktop, Featured