• OpenStack, Outer Space & Left Eyes

    I just realized that I have a conflict of abbreviation.  Common interpretation of “OS” in the industry is generally operating system, and sometimes open source.  To those outside the cloud industry, perhaps one-size, outer-space, or even oculus sinister (the left eye).

    Not to minimize the importance of an operating system, a left eye or outer space, but for the sake of saying:  I hereby propose that the old-skool meanings of OS are obsolete.  Used in a naked context, or without clarifying qualifiers, the OS acronym shall now equate:  OpenStack.  Granted, I could be biased.  Just a silly thought this morning, there you have it.

  • Offline Instance Migration with Nova Compute

    In managing or testing an OpenStack private cloud environment, it may be periodically necessary to offload nova compute workloads to perform planned server maintenance. Live migration is often preferred and is generally the best option, as it requires zero instance downtime. Motivations for choosing to do an offline migration over a live migration can certainly vary, depending on any number of variables such as shared storage configuration, hardware condition, hypervisor flavor, or perhaps the nova compute, qemu or KVM versions that are in play. Whatever the reason, here is a simple example of an offline migration.

    This environment is OpenStack Icehouse 2014.1, Juju-deployed onto MAAS-managed Ubuntu Server 14.04 LTS Trusty Tahr nodes. Storage back-end is ceph RBD, but other shared storage options should do as well. It’s all open source, and it’s all free as in … freedom to learn, or freedom to buy support if or as needed. By the way, I just love that about open source. Ok back on track, here goes:

    The short version:

    # For offline migration, the instance should be
    # in a STOPPED / shutoff state!  Use nova stop for
    # non-graceful shutdown of the vm, if you must.
    
    # Get the instance ID
    nova list --all-tenants | grep my-instance
    
    # Confirm the current state and hypervisor node
    nova show ab90cc7a-7022-4157-b32a-11c49b7701cc
    
    # Migrate the instance, nova scheduler decides where
    nova migrate ab90cc7a-7022-4157-b32a-11c49b7701cc
    
    # Watch & wait
    watch nova migration-list
    
    # Seal the deal
    nova resize-confirm ab90cc7a-7022-4157-b32a-11c49b7701cc
    
    # Confirm the goods
    nova show ab90cc7a-7022-4157-b32a-11c49b7701cc
    
    # Fire it up
    nova start ab90cc7a-7022-4157-b32a-11c49b7701cc

    [Read More…]

  • OpenStack One-Liners

    One-line bash commands for managing, deploying and testing OpenStack and related components, hypervisors and containers.  Some simple and obvious, some cool and questionable, hopefully all useful.  Check back on this post, I’ll likely add more.

    #1 – Watch Nova Compute for Non-Active Instances on Openstack

    Outputs total instance count and non-active instances. Useful for keeping an eye on potential problem VMs in a PAUSED, ERROR, SHUTOFF or other non-active state.

    Command:
    watch --interval=1 'nova list --all-tenants | wc -l && nova list --all-tenants | grep -v "ACTIVE"'

    nova-tenant-list

    [Read More…]

  • Joining Canonical: Ubuntu Server & OpenStack Team

    ubuntu-for-rbwordpressFor over 15 years, I’ve been leveraging the power of Linux and Open Source Software in business network deployments.   Through my former consulting company, and eventually the company which acquired that business, there have been many cherished experiences, and countless invaluable relationships forged.  The time has come to change direction from the managed services, business consulting and system deployment worlds, over to the planet where such Linux things are born.

    For the past 4 years, I’ve worked with the business and technology minds at SCTelcom to combine and transform my former business into the area’s premier IT solutions provider. Together, we have fine-tuned a serious team of talent to support this area’s businesses and our own ISP/Telcom operations. I am very confident in SCTelcom’s continued commitment to provide solid solutions, responsive support and top-notch talent. The company is strong and it is dedicated to the people and businesses of this area.

    canonical-250-for-rbwordpressNext week, I join Canonical on the Ubuntu Server and OpenStack team.  The team I’m about to join is a global group of developers, unified on the mission of producing the best open source platform in existence.  Working for Canonical will be a new culture and an exciting experience with a lot of new ground for me to cover.  Conversations with Canonical staff over the past couple of months have given me a glimpse into a work force who are each extraordinarily passionate about their work in open source.  I will no doubt miss my former coworkers, employees and long-time clients, though I know that they will be in very good hands with SCTelcom’s talented crew and bright future.

    Jenkins, JuJu, Launchpad, Ubuntu LTS, UDS, OpenStack Summit and work item burndowns – here I come!

    A big huge THANK YOU to the area businesses and individuals who have been a part of my business & career over the past 15 years. I am so grateful for the support, patronage and friendship! #blessed

     

  • Moving from CentOS to Ubuntu

    It’s no secret that I have historically been a RHEL and CentOS guy.  Red Hat Advanced Server before that.  I think the recent bond made official between Red Hat and CentOS is exciting for those groups and the affected communities.  Over the past several years, I’ve seen and heard from peers and others the industry … a massive excitement over Ubuntu.  But I already had my stable platforms humming along, so I didn’t pay much attention to Canonical or Ubuntu.  Or at least not until OpenStack.

    For anyone wanting to deploy OpenStack, Ubuntu is the reference platform to do so.  RHEL and CentOS can do it, but it is currently just not a natural fit.  Canonical has one heck of a head start with JuJu, MaaS and Landscape.  Granted – one can indeed use packstack or RDO on RHEL-based distros.  That is still a long shot from where Ubuntu already is today.

    When I decided to seriously consider a career shift in working for Canonical on the Ubuntu Server and Openstack Team, I rebuilt my business and home stacks with Ubuntu LTS.  Prior *nix experience made for a quick pick-up of the Ubuntu/Debian similarities and differences.  In the interest of not inviting yet another Linux distribution debate, I’ll just say that I quite like Ubuntu LTS.  Ubuntu server is now my platform of choice.

    What about Unity?  Mir?  Upstart?  I’m traditionally an X control freak, and enjoy XFCE’s ability to be tweeked and tugged how I see fit.  When I rebuilt my desktop powerhouse, I stuck with XFCE on Ubuntu LTS.  But on my new work laptop, Unity it is!  It just works, no manipulations necessary.  Linux for human beings, check.

    The asynchronous Upstart is probably my biggest acceptance challenge at the moment, but I attribute that to not possessing granular knowledge of it yet.  I understand the contrast and benefits, but that hasn’t stopped me from occasionally tripping over my init tendencies.

    Canonical’s goal of a unified code base for server, desktop, mobile and others is quite a massive and worthy undertaking.  While it may seem to many that they are reinventing the wheel with Mir, and that they have already reinvented the wheel with Upstart, these bold moves are what set Ubuntu aside from all other distros.  The community and the company are investing in some fairly major changes.  Change always brings stress and resistance.  Pushing through the uncomfortable and awkward beginnings is how change is made successful.

    Push on, push on!

    ——————– ——————– ——————–

    Addendum – a useful Ubunto Wiki doc for those making the same journey.  After completing a recent shift from RHEL/CentOS to Ubuntu, I found this nice reference:

    SwitchingToUbuntu – Switching from Red Hat to Ubuntu;  Switching from CentOS to Ubuntu;  an equivalent command translation reference.

    • chkconfig vs. updaterc.d
    • service start vs. invoke-rc.d
    • yum search vs. apt-cache search
    • yum install vs. apt-get install
    • rpm vs. dpkg

     

  • CentOS 7 Beta and RHEL 7 Beta

    It seems like we just made the transition from RHEL and CentOS 5.x to 6.x.   Our 6.5 boxes are running smoothly and…  here comes EL beta 7!  Red Hat announced RHEL7 Beta availability on Dec 11, 2013.

    Karanbir Singh and the CentOS community are posting news, views and reports about CentOS 7 beta downstream here:  http://seven.centos.org.

    I plan to have a box online shortly just to see if RHEL7 / CentOS7 plays well or doesn’t play well with OpenStack.