With the 6248 fabric interconnects they follow the path that the Nexus 5K series from Cisco have gone with all the ports being Unified Ports. Meaning they can do 1/10GigE or Fibre Channel. I like this direction because it offers more ports and greater flexibility for the FI’s and customers. But when it comes to configuring the FC uplink ports it’s not as clear as it was in the past. You would think that if you right clicked any of the ports you would be presented with an option to make the port an FC port but that’s not the case.

Below I am outlining the steps it takes to enable ports for FC uplink status.

Step 1: You will make sure you are on the equipment tab, then click on the first Fabric Interconnect. In the window to the right you will see the following options in the Actions area. You will need to click on the Configure Unified Ports option.

Step 2: You are going to see a window open that matches the one in Image 2 below. This shows a visual presentation of the ports in your fabric interconnect. The most important thing here is the somewhat obscure white slider bar that is just below the image. I have pointed an arrow to it. When you first arrive at this screen the slider will be all the way to the right, I have slide it 2 rows to the left. This will allow me to configure the last 4 ports as FC uplinks. You can adjust for what you require for your UCS design.

Image 2

Step 3: In image 3 below I have placed a box around the ports that I intend on changing to be FC Uplink ports to make sure it’s clear.

Image 3

Step 4: The last thing to do in this part of the config is to right click on each port and choose Configure as FC Uplink Port. This will allow you to connect to your switching fabric.

Image 4

Step 5: After doing this on the first FI it will tell you that it needs to reboot. Once the FI has rebooted you will need to do this same config change on the 2nd FI.

Step 6: Now when you look in the tree and find the section for FC Uplinks you will see the ports that we configured. Now make sure the proper VSAN is enabled for each port and you should be ready to proceed with the rest of your install.

I want to say thanks to my coworker “Mr UCS” Steve Pantol for the heads up on this saving me time trying to figure this out.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

I am not surprised by the lack of prep material since it’s a design based exam. In my opinion this is one of the exams that you just need the real world experience of working with the product over time to be confident that you have the knowledge to pass the test.

Of course I’m not going to spit out a bunch of questions that you should study. The test does present a bunch of different design scenarios that might cover things like environmental variables, security questions or technical requirements. You then must make your choice based upon the given parameters. The part that I was least happy with was how little content there was about actual UCS design decisions. Sure there were things like converged networking and cabling, but actual questions that required you to build a design were somewhat limited. The exam seemed to focus more on the overall Cisco data center methodology rather than just UCS as you would think.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

Cisco UCS Manager – Does all the following in one tool

• Blade Chassis mgmt
• CIMC console connections
• Hardware Monitoring
• Firmware updates – chassis, FEX modules, blades and adapters
• QOS
• Network Mgmt
• Service profiles for blade identities
• SAN connection setup

Roll Based Access Controls – Can assign ID’s access to all or just allow server, network or SAN access

HP Blades – using Virtual Connect modules of any flavor

• Blade Chassis mgmt. – Onboard Administrators (OA)
• iLO console connections – OA
• Hardware Monitoring – HP SIM
• Firmware updates – HP SIM, FDT or currently used tool
• Network Mgmt – Virtual Connect Manager (VCM)
• Server profiles – Virtual Connect Manager (VCM)
• SAN connection setup – Virtual Connect Manager
• Roll Based Access Controls (RBAC) – need to configure logins per Tool

If you elected to use the new Cisco FEX modules for HP Blade chassis then the following would be affected.

• Network Mgmt / profiles – no profiles, would use physical address from Blades. Port configs done at 5K level. Blades auto map to ports on FEX similar to pass through connections.
• SAN connection setup – Would be done vid FCoE through the 5K switches
• Network QOS – Done on 5K switches and tagged if possible in Operating Systems (ESXi 5 does support 802.1p)

If you have anything that I left off of this list drop me a note in the comments and I will update to keep accurate.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

This is a big bonus of working Cisco UCS because there is just one place that I need to go to configure, manage, patch a blade environment. With HP I might have to use the Onboard Administrators, Virtual Connect manager and HP SIM. This seems like such a no brainer that HP could have fixed years ago, because the HP Blade systems have been around for years.

So if HP wants to rule the market in the Cloud era they need to reduce their tools into less management points. I should be able to do everything in one console for compute and if they could roll in storage to this also would be a huge win. And if they are properly motivated they can. Now a unified tool for automation of hardware that could reach into VMware with API’s would be the next big win. I know they are working on some of this but last that I saw it still sounded like there would be too many moving pieces.

So HP I challenge you to solve this issue with your massive amount of Talent, code and other resources. You need to become innovate again and make products that people want to own and desire for other reasons than you have a large install base and are cheaper.

If you want some examples you can read my post about the comparison of what tools it takes to manage HP vs Cisco blades chassis here.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

For example if you are running ESX(i) on a mirrored pair of local drives, if you are a shop that does not have very good hardware monitoring you might have no idea of the health of this mirror that vSphere is running on. So this becomes even more important with more shops experimenting with running certain workloads on local disk. With VMware and storage companies creating Virtual Storage Appliances that can run on these local disks and still provide the benefits of shared storage, this becomes a must to understand what is happening in your local disks environment.

With the latest batch of CIM drivers from HP they are now exposing some of these details. You can now see the drive configurations and status. The image below shows that a drive is rebuilding in the RAID config. This should be a feature that many HP shops will be happy to see. If you have noticed any other good features from this update leave a note in the comments for others.

a

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

In the end its now working as expected and is a great feature to put to use. With so many different pieces of equipment in your environments being able to have a unified log in is much better than trying to remember 20 different local ID’s.

You can download and view a copy of the LDAP for Cisco UCS guide from here or a Google search will turn up the same thing.

Error #1

In the “Creating LDAP Provider” section the main part that tripped me up was the following.

If you refer to the image below shows the highlighted portion according to the document. I have updated the instruction below. Once I changed this the authentication worked immediately. Before changing it would just fail and according to the logs on the AD server it was not even making the attempt and failing.

c) This should be the string for the Bind user that you created earlier in the document. Example below

BindDN value is CN=ucsbind,OU=CiscoUCS,DC=sampledesign,DC=com

Error #2

This section in the collecting information section was also wrong. It did not cause me any issues but did require me to go back and read things a few more times to make sure.

In part d it references OU=CiscoUsers in the string. But the instructions never requested us to create this OU. It should just be the OU=CiscoUCS that you did create. Nothing to cause you issue just to clear things up.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

The Windows 2008 x64 image being used for the basis of the install for the application servers (vCenter, Fabric Manager and Netapp utilities server, etc) is very security focused, and as such, can be a real pain for application installs.  In this case, for Fabric Manager (version 5.0.4b), we did the following:

1)      Unzip Fabric Manager, and double click your way into the software folder

2)      Install Java from jre-1_5_0_11-windows-i586-p.exe by right clicking and choosing Run as Administrator.  One thing of note:  this was a fresh install and did not have an existing install of Java.  As I’m sure we all know how fussy Java can be, I would recommend this being installed in a location that does not require any additional Java-oriented applications

3)      Once installed, I had to put the java bin folder into the PATH for the server, else I could not get the Fabric Manager installer to run correctly.  (this is the path: C:\Program Files (x86)\Java\jre1.5.0_11\bin)

4)      Next, head into the postres folder (located in the software folder, same as the java installer), then the Windows folder and run postgresql-8.2.msi. This will install the Postgres database that will be used for Fabric Manager.  During the install, we accepted the default components, then set the password for the service that it would run as (you can choose the name, the default is postgres).  This will also create the account for you as part of this process:

5)      Next you will be able to give the superuser account for Postgres a password, and remain the account should you want to.  You should also set the locale (English, United States in our case):

6)      The procedural languages selection is fairly straightforward, as only PL/pgsql is available as a choice.

7)      For the enable contrib modules, I leave the default selected (Adminpack.)

8)      The install should continue and complete.  When it is done, we need to now create the database that will be used for Fabric Manager

9)      Launch pgAdmin III from the start menu:

10)   On initial launch, you will see it the admin gui is disconnected from the default postgres instance.  Right click and choose Connect, then enter your password you created during install:

11)   To create a database for Fabric Manager, right click on the default instance and select New Database:

12)   You can name the database whatever you prefer, but make sure to set the Owner to the user postgres:

13)   Now it’s time to launch the fabric manager installer (finally, right?).  You can do so from the start.html .  Don’t worry if you get an unsupported environment pop up, just hit okay and choose a Custom install

14)   We went with the Standalone install, and when the Database Options appears, make sure to select Use Existing DB, set the DB superuser account (postgres by default) and the password that you set during the install for Postgressql, and set the DB URL to have the name of the database you created in place of “dcmdb”

15)   Set the local fabric manager user (defaults to admin, we used fmadmin) and set the password:

16)   For the Authentication Options, since these are standalone installs, the mode we are using is Local

17)   And for Configuration Options, we check both the “Use FC Alias as fabric default” and “Require SNMPv3 and disable SNMPv2c for increased security”.

At this point, you should get the Fabric Manager and Device Manager icons on your desktop, and you should be all ready to go.  Hopefully this will help should you try the Express install and run afoul of the many issues of security that can come up during application installs.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

The servers came with ESXi 4.0 installed on a USB stick from the factory and installed in one of the two internal USB ports that the server offers. Upon turning on the servers some of them booted right to VMware and some did not. After some further looking into the boot order in the BIOS I noticed that the Embedded Hypervisor option was not added to the boot order on a couple of the servers. A quick add and they were running just like the rest, guess someone at the factory missed that one.

The servers took a very long time to post and boot up, part of this was due to the 128 GB of RAM installed. We turned off some of the non-essentials and modified the boot order to go right to ESXi and cut the post time down some. You can see from the image below it’s just another x-series server.

I snapped the image below with the cover over showing off all the sticks of memory installed.

The last image below is a close up to the two USB ports that are internal to the server. The lower one as the USB stick from the factory with ESXi embedded on it.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

I will be creating some posts myself over time about these same features.

This next video is walking you through the creation of the Service Profile and explains the ability to create an Updating Service Profile that will update child templates created from it. This would help you keep those profiles all in compliance with your changes.

The next video talks about creating your UCS service profiles and pre-provisioning servers before the blade servers arrive. This method allows for you to do the work up front then when the blades arrive just install them and your work is done already.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)

So I figured that it would be good just to write down a few of the little things that have impressed me so far. I will be writing a lot more about UCS in the coming weeks. But these are just some UCS features that I thought were cool.

This is no surprise but does the back of your server rack look this clean? Unless you have a UCS blade chassis I doubt it does. Sure other vendors have been creating Blade Chassis for years and they have done many things to cut down on cable clutter. But nothing comes close to making things this simple and clean.

The next one is maybe not so much a technology innovation but it’s just something so simple that I can’t believe no one has done this before. On each UCS blade server that is a little paper card that flips out. This can be used to write server names, put asset tags or other labeling details. No more are the days were you are forced to paste labels on the front of servers reducing the air flow by partially covering up some of the vents. This seems so dang easy but I’ve not seen any other vendor do this yet.

This will probably have people split on if its good or bad. Every UCS blade and C series rack mount server has the console port on the front and you can use the dongle in the picture below to access. The UCS dongle provides you with a video port, 2 USB ports and a 9 pin serial connection. This gives you the ability to connect monitor, keyboard and mouse to any blade or server. You could also use it for a console connection to a nearby switch if your laptop like many does not have a serial port. Sure others will probably say why would you want this when I just cable up my chassis to a KVM and forget about it. But after years of working with remote data centers and having a wide variety of skilled and non-skilled works there to be your hands in a crisis. This makes things dead simple just connect this dongle to server 1 and what do you see on the screen. No more try to remotely talk someone through how to use a KVM and never really being sure if they are looking at the right screen.

About Brian

Brian is a Technical Architect for a VMware partner and owner of this website. He is active in the VMware community and is helps lead the Chicago VMUG group. This blog Virtualize Tips was started to document and remember things that I come across while working with tech.

Mail | Web | Twitter | LinkedIn | More Posts (170)