Recent Updates Toggle Comment Threads | Keyboard Shortcuts

  • idid 11:10 pm on January 10, 2012 Permalink | Reply  

    Configuring iSCSI Boot in Cisco UCS Manager 2.0.1 

    You’re going to make these decisions (or get the information):

    A Blade with an M81KR (VIC) * I don’t have a Broadcom Adapter to document

    Chassis and Slot ID for that Blade

    Service Profile Name

    UUID (or Pool to use)

    Adapter Policy (Linux, ESX, Solaris)

    A MAC address “preference” (you can use a pool if you prefer)

    Storage Target IP (a NetApp or EMC device configured for iSCSI access)

    iSCSI IQN (iSCSI Qualified Name) of that Storage Device

    The Primary Initiator’s IQN (iSCSI Qualified Name)

    The Secondary Initiator’s IQN (iSCSI Qualified Name)

    The Primary NIC’s Initiator IP Address

    The Secondary NIC’s Initiator IP Address

    IP Gateway Address for the subnet being used

    DNS Server IP address for that Subnet

    Subnet Mask

    Subnet VLAN configured elsewhere in UCS Manager and given a “Name”

     

    Here is a table of the decisions or data from my Lab Environment:

     

    image

    My first pass at documenting will just be the screen shots of the UCS Manager after I implemented the “answers” to the “Expert Service Profile” wizards. I will use this post to re-create the Service Profile and snapshot the actual screens in a subsequent post.

     

    I created a Service Profile called “ISCSI24” (with 24 implying Chassis = 2, Blade Slot 4.

    I also input the “24” scheme into the UUID field manually entered.

    image

     

    This is the bare essential NIC configuration: 2 vNIC’s to be mapped for iSCSI use and 2 vNIC’s that should be used for public networking on the to be installed OS. It makes sense to implement the higher numbers for iSCSI so the public Networks end up later being eth0, eth1..

    I alternate the NIC across Fabrics A and B and hand-coded the MAC addresses slipping “24” into the fields and numbering them sequentially to match the vNIC number (00, 01, 02, 03).

    image

    New to the iSCSI feature of UCS Manager 2.0 is the ability to configure iSCSI vNIC’s.

    They must be associated with one of the configured vNIC’s above. I chose vNIC’s 2 and 3.

    If your target OS is VMware you’ll want a lot of vNIC’s and use the last 2 for iSCSI overlays.

    The “default” iSCSI adapter policy is recommended for the VIC. The other options are to implement iSCSI “offload” features of the Broadcom adapter.

    image

     

    When you create the “Boot Order” configuration there’s a new iSCSI vNIC section. Add the CD ROM for OS installs and then insert the 2 iSCSI vNICs.

    image

    That little “set Boot Parameters” widget on the bottom left in the above image is key: it pops up a selection widget for iSCSI0 and iSCSI1 and any others you might implement. Each gets configured in turn and this is where the key iSCSI configuration data is entered.

    image

    The “Boot Parameters” have all the key fields: Initiator and Target IQN’s, IP addresses and Network/Subnet specifics. This service profile can be duplicated to boot multiple LUN’s from the same iSCSI target and differ only by the Storage LUN requested.

    image

    The data for ISCSI1 (the alternate boot path) to the array:

    image

    On the NetApp Array you must configure the UCS Blades Initiator IQN’s into an ISCSI  Initiator Group

    image

    There’s an ISCSI Report that shows the Array’s IQN (and Alias) and specific NIC’s enabled for ISCSI access with their IP addresses.

    NOTE: I haven’t tested using the Alias or changing the default names provided on the NetApp (yet).

    image

    You must create a LUN of adequate size and then MAP that LUN to the ISCSI Group you created (ISCSI24 for me) and configure the LUN ID that will be configured on the UCS side.

    I created LUN 0 for a Red Hat 6 install. and LUN 1 for ESX 4.1.0 Update 1 (the only supported OS’es for iSCSI Boot) that I care about… yes Windows version are supported. Not my thing.

     

    image

     

    That should do it… assign the *new* Service Profile to a Blade and bring up the Blade’s KVM Console. When you see the big Cisco logo after mutilple passes through the PNUOS Utility OS to install all your configuration options you need to Hit “Escape” to see the VIC ISCSI configuration “test”. If everything is not aligned correctly you’ll see this “Error” that flashes up for 2 seconds:

     

    image

     

    If everything is correct you’ll see information about the LUN discovered on the Array.

    I hope this helps someone save the hours of “Trial and Error” I spent understanding what was required in each of the configuration fields… I was close… but I saw the image above over and over until everything was just right.

    I found the essential missing clue by Googling the error message and finding a “Zip file” of screen shots similar to what I have posted here. I’m sure a lot of folks can’t just download Zip files at work.

    Google search for “Cisco VIC iSCSI, Boot Driver Version 2” + “Initialize error 1” and you’ll see the Cisco community forum reply with the alternative Zip file of images from Simon Geary… Thank you Simon for the pictures worth a thousand “key strokes”.

     
    • Craig 6:33 pm on March 20, 2012 Permalink | Reply

      What code version was this? 2.0(t) on the FI’s? I am running 2.0(w) and don’t have the “Set Boot Parameters” under the iSCSI vNIC’s. Am able to config same stuff, but per iSCSI vNIC but not as shown in your documentation. This is all pretty kick ass, thanks for your work.

      Craig

    • Craig 11:06 pm on March 20, 2012 Permalink | Reply

      think the Cisco VIC iSCSI, Boot Driver Version 2” + “Initialize error 1 happens when two initiators try to login to same target. First makes it, second fails.

    • las vegas Bail bondsman 1:11 am on November 8, 2012 Permalink | Reply

      Nice post. I was checking continuously this weblog and I am impressed!
      Very useful information specially the remaining part :) I handle such info much. I used to be seeking this certain information for a very lengthy time.
      Thank you and best of luck.

  • idid 9:38 pm on January 7, 2012 Permalink | Reply  

    McAfee warns me that Dave Winer is dangerous… as if I didn’t know already. That’s why I read him… to crank up my brain. 

    scripting_new_is_flagged

     
  • idid 9:28 pm on January 6, 2012 Permalink | Reply  

    Daily Reads 

    http://theimmoralminority.blogspot.com/

    http://andrewsullivan.thedailybeast.com/

    http://littlegreenfootballs.com/weblog/

    http://talkingpointsmemo.com/

    http://www.huffingtonpost.com/

    http://www.thedailybeast.com/

    http://news.ycombinator.com/

    http://scripting.com/

    http://slashdot.org/

    http://planet.intertwingly.net/

    http://siliconangle.com/

    http://www.memeorandum.com/

     
  • idid 8:38 pm on December 29, 2011 Permalink | Reply  

    iPad Email Signatures? 

    I see there’s a text box for adding a signature for out bound email on my iPad but is there a way to embed an image (a company Logo in this case) into the text and have it loaded into the email?

    RESEARCH STARTS… Update to follow

    Arg. There’s an app with the feature but you must send the outbound email through it.

     
  • idid 9:11 pm on November 25, 2010 Permalink | Reply  

    Why are there 7 white and 5 black keys per octave on the piano? 

    A comment I posted at:

    http://math.stackexchange.com/questions/11669/mathematical-difference-between-white-and-black-notes-in-a-piano/11873#11873

    The math of frequency relationships here is sound (pun intended) but they don’t help explain the white vs black key piano layout.

    Here’s the historical imperitive that led to this layout for “Western Music”.

    First consider the major triad:

    root + third + fifth notes of the “diatonic scale”.

    They follow the harmonic series:

    1. root
    2. octave (doubling of root frequency)
    3. fifth (triple of the root – 3:2 relationship to the octave)
    4. double octave (4x)
    5. 10th (double octave of a third)
    6. octave fifth

    The Bugle notes: These are notes a static length of tubing can produce by blowing into it and producing standing resonant waves of 1x, 2x, 3x, etc.

    Combinations of these notes create frequencies that make choirs sound heaven-ly. The frequencies align and blend into pure complex vibraations that are the sum and the differencies (harmonic overtones) of these relationships.

    Choirs can tune themselves dynamically to create these frequency alignments that are percieved as being perfectly consonant. Upbeat western music focus on the 3 major chords found in the diatonic scale:

    • 1+3+5 root major chord – white keys C – E – G 4+6+1
    • 4th chord – white keys F – A – C 5+7+2
    • 5th chord – whote keys G – B – D

    The basics of western folk music are the 1 – 4 – 5 sequences of chords. Learn C, F and G on a guitar and you can play the bulk of the classic Country song book.

    Put the notes of these chords into a scale and you get that row of 7 white keys: C – D – E – F – G – A – B (repeat until you can’t hear it).

    So, they western scale is based upon frequency relationships that make combinations of notes “ring” in consonance in it’s purest form… like the Gregorian Chants of the Roman Church.

    So, a basic “western keyboard” could be made from just these 7 notes repeated across the frequency spectrum. Look at the layout of a Greek Lyre (a harp) and that’s what you will find. A sequence following the diatonic scale which sounds pleasant if you just strum across the strings due to the tuning of even multiples (adjusted by octaves).

    OK… now adding the black keys is a compromise of tuning specific notes so that you can build these 1+3+5 chords from any starting point and thus play a song adjusted up or down to any starting point. The piano will never achieve that sonic mathematical glimpse into the “music of the spheres” that the self-adjusting choir can to make a chord mathematically perfect in alignment but it’s the “keyboard” for the modern composer… the effective “musical qwerty” that a composer or a pianist begins to visualize chord “shapes” as hand positions.

    With a lot of practice a pianist can pre-visualize sound in terms of finger and hand movements much like a solid touch typist starts to set words and sentences as a sequence of movements.

    The addition of the black keys was called a “Well Tempered” tuning and Bach was one of the first composers to create whole bodies of compositions that worked through the Major and Minor keys of the 12 scales that you noticed intially when inspecting the keyboard.

    If you look into other musical cultures you will find a different approaches to standardizing sound relationships that do not focus on the 1 – 4 – 5 chords. This music to a culturally trained western ear is less predicatable in nature and that lack of predictablility can make the music frustrating or exciting… music “speaks” to us in terms of pure sensory inputs that can move, excite, bore or confuse us.

    So, the piano keyboard is designed to be the perfect delivery system for an individual to produce the range of complexity that western music has achieved.

    The modern keyboard synthesizers are now able to produce the full range of the western orchestra in terms of “instruments” and I’m hoping someone create one that micro-adjusts notes based upon the surrounding context… shifting a note up or down slightly from the “well tempered” compromise to the pitch that makes a chord “ring” and produce the upper harmonic overtones that make a great orchestra truly “heavenly”.

    Maybe it’s already been done.

     
  • idid 5:39 am on March 10, 2010 Permalink | Reply  

    Jonathan Schwartz tells some IP Patent stories and how Sun responded to Apple and MS on his blog 

    IP patents are useful defensively and offensive when used to inhibit competition. Or “license” the efforts of a competitor. I will hold off on any Apple purchases until further notice.

    Posted via web from mcd’s preposterous

     
  • idid 6:06 pm on February 3, 2010 Permalink | Reply  

    Re; iPad-like devices – “I think calling them computers is wrong”. Yes. Call them Slates. 

    I think calling them computers is wrong

    Call them Slates. Does anyone own that word yet? HP maybe.

    Posted via web from mcd’s preposterous

     
  • idid 6:00 pm on February 3, 2010 Permalink | Reply  

    Bingo: “tomorrow’s computing systems, heralded by the iPhone (and iPad), are not for DIYers.” 

    But tomorrow’s computing systems, heralded by the iPhone (and iPad), are not for DIYers.

    Zeldman speaks pure geek genius. Adding plug-ins (like Flash, Java) to fixed function “computers” makes them less effective… and it damages the ability of search engines to find content.

    Now if we just had a license free format for video… we’d be close to an internet that it’s “owned” by companies and rented to us (with overlords).

    Posted via web from mcd’s preposterous

     
  • idid 3:33 am on February 2, 2010 Permalink | Reply  

    “I don’t welcome my Apple overlords. But I welcome the change, at least for now.” – a geek must read 

    I don’t welcome my Apple overlords. But I welcome the change, at least for now.

    Posted via web from mcd’s preposterous

     
  • idid 3:27 am on January 31, 2010 Permalink | Reply  

    The iPhone does have Google Street View maps! Seriously. I’d never have found it without a clue. 

    Go and read it all. Then this picture makes sense to you.

    Posted via web from mcd’s preposterous

     
c
Compose new post
j
Next post/Next comment
k
Previous post/Previous comment
r
Reply
e
Edit
o
Show/Hide comments
t
Go to top
l
Go to login
h
Show/Hide help
shift + esc
Cancel
Follow

Get every new post delivered to your Inbox.