RFC 2119…

Please keep in mind that the key words “MUST”, “MUST NOT”, “REQUIRED”, “SHALL”, “SHALL NOT”, “SHOULD”, “SHOULD NOT”, “RECOMMENDED”, “MAY”, and “OPTIONAL” in this document are to be interpreted as described in RFC 2119.

One of my peers cited this in one of our project meetings.  Nebulous or poorly interpretive descriptions of technical requirements should be avoided.

\\ JMM

Well engineered projects are indistinguishable…

From:  http://www.dslreports.com/forum/r29293609-How-Does-WHT-Do-It

“Well engineered projects are indistinguishable from crazy ideas.”

I resonated with this statement as we spoke of the important of wiring standards, craftmanship, and the sad state of installations in so many IT shops.  Why are well engineered projects, like data center wiring, met with so much resistance?  Answer: cost, time, and effort are not always well articulated nor understood by the masses.  And often, trump all unless you spend the time showing amazing value for that effort.

A wire is not always a wire.  Kudos for those shops that get it and embrace good IT.  It’s crazy, I know.

\\ JMM

Think like a Risk Manager…

“While I understand the process and how it should work, there is a chance that someone could go in and make changes [to servers]. We have to think like a Risk Manager and the possibilities that could happen.”
– Steve Moore, Director, IT Operations, Santander Consumer USA (2017)

Just recently, we had several conversations where system engineers lamented on the amount of work risk mitigation has created.  While this often is viewed through various colors of lenses and often tempered with bias, the point was not to just express exasperation about the volume of reactive work.

The point was to proactively think like a risk manager and head things off so it’s built into the DNA of the technology.  Are we really thinking this way?  Are we creatively thinking about risk as we architect solutions.

Let’s prevent the backlog versus react to it.

\\ JMM

The Year of KNOWLEDGE…

From:  https://community.servicenow.com/events/2951

“It’s Official, 2017 has been coined “The Year of KNOWLEDGE”. Many, if not all of you, have started, or plan to start your Knowledge Management Initiative this coming year.” – Josh Addington

It’s probably no surprise that managing our proprietary and intellectual knowledge for commodity services, such as technical support, is still a problem in 2017.  Interestingly, people are doing something about it through community initiatives.  This is one such here in Dallas, Texas.

Excited to see what fruit this will bear, what ideas can be shared, and if we must, collectively display our sorrow at the state of our own challenges in this tough space.

\\ JMM

Lack of intuitiveness drives complexity…

“My feedback is the lack of intuitiveness drives complexity.”
– J. Merrill

The context of this quote covers so many different areas, in my career.  Everything from user interface and workflow discussions to policy and procedure brainstorming.  It’s very easy to run simple into the ditch.

Simplicity can only be accomplished with the addition of intentional and intuitive interfaces in writing, electronic, and in practice.

\\ JMM

What A Next Generation Network Monitoring Solution Looks Like…

Many vendors tout “Next Generation Monitoring” solution, yet upon looking, looks like what I’ve seen for many years.  Having had a few tough discussions with sales people, the next generation moniker is quickly becoming a sales tag line and nothing really disruptive to the market.  In today’s market, considering DevOps and SDN, tools are far more important today for doing more with less people.

If you’re selling a network monitoring solution and feel your solution is next generation, please read.

Core Requirements

Business Intelligence Driven

  • Meaningful, amazing, action compelling reporting. Most canned reports are lame and don’t add value.  Give IT Managers and System Engineers reports that are incredibly insightful.
  • Create fear… Show people how bad performance of the physical network, Active Directory, Exchange, and SQL environment is… Shock or affirm me.

AI-Driven Discovery, Identification, and Monitoring

  • Manually defining hosts and services is so 1980s… NGNM tools discover what is out there, where it is, and give visibility to what should be monitored.  Unleash the tool and let it do the work.
  • Leverage AI to determine what things are. Manufacturer recognition, SNMP and WMI. Profiling works. Apply the concept here.
  • Leverage the cloud by providing the database centrally. Don’t make me track down SNMP Mibs.
  • Go beyond hosts and MIBs. Monitor IP Addressing (IPAM), Storage platforms, and cloud services.

Business Views, System Views, And 360 Views

  • Include the physical datacenter. 2D/3D model of the datacenter, what’s in the cabinets, etc. Take what is discovered and place it in this vide.
  • System views. Dynamically create core infrastructure views: LAN and WLAN. But also Active Directory, DNS, DHCP, Replication, SQL, SharePoint, etc. Identify unknown servers and services, forcing Engineers to get involved and document what is out there.
  • Business views are good ways to see how systems interoperate, but affect the whole. WAN goes down, this is what it effects. Especially important when LAN meets Cloud services.

Intelligent Configuration Change Management

  • If your scanning anyway, alert on changes to the environment. The tool needs to be able to fire alerts when they see a change from point A to point B.
  • Connect to Change Management systems, like ServiceNow or ServiceDesk.

Workflow Automation

  • Alerts trigger actions.  Open a ticket.  Run a script.  Stop and restart a service.
  • Virtualization automation.  UCS automation.
  • Or offer to plug into MS SCCM or VMWARE Orchestrator.

User Experience (UX)

  • Clean, object based, tablet friendly user interface. Tabbed interfaces are great, if done smartly and intuitively.
  • Use tried and true web UI navigation, such as breadcrumbs. Should take no more than 3 clicks to get to pertinent data.
  • Dashboards and core technology modules should be modular, configurable, and reset-able.
  • Adding URL’s or jump offs by host. NGNM says, “This server is running Splunk and here is the jump off.”

Documentation Repository

  • Documenting systems is a major problem in the majority of IT shops. The NGNM should begin to leverage what it is gathering and offer to put together the documentation.
  • Provisioning documentation and configuration snapshots (Check outhttp://sydiproject.com/ to see a starting point).  NOC should be able to leap off the site to where the docs are.
  • Change Management “changes” should be reflected in documentation.
  • Give me something I can print. PDF preferable. Something I can give auditors.

Education & Community

  • How do people spin up on the NGNM? Wiki is good, but there are better ways to educate and sell value.  For example, YOUTUBE.  Show me how to win.
  • An active community full of ideas, helping each other, examining use cases, and growing the influence based on wins.  Include me into a community of people wanting to win.

\\ JMM