The Scenario: The Can-Do organization is a heavy construction equipment leasing headquartered in Madison, CT and with additional operating facilities in Virginia and Nevada. Their data centers are located in Madison, CT and Richmond, VA. To support the leasing business, they also have a credit company that has evolved into a full-service bank and insurance underwriter and broker located in the firm’s 600 retail centers. Banking applications were housed in the Richmond, VA data center.

The Plan: The plan we were given was called the “Emergency Response Plan” dated July 15, 1998 and prepared by the VP of Training, Cafeteria Services and Vehicle Maintenance. HA-HA.

Recovery teams included Operations, HR, Finance, IT and physical security. I was on the IT recovery team.

Although the plan stated that Can-Do ascribes to the U.S.’s National Incident Management System (NIMS), no one knew who was in charge of the incident nor how to communicate with the incident management team. It was even questioned as to why a command center was stood up on the first day of the hurricane warning; that foresight proved to be fortuitous as it turned out. The first thing we all needed to do was elect an incident commander (assigned to the Operations recovery team) and one for each recovery team. All recovery teams also assigned coordinating responsibilities to their sister recovery teams. The IT recovery team was smart (wink wink) and assigned a supply chain/vendor management coordinator and a scribe – guess who took that role (me).

We started the exercise by getting a hurricane warning for Madison, CT along with a warning that a mild flu pandemic was possible. The first step that the IT team performed was topping off the fuel tanks for the CT data center generators. We then sent out a message to the IT staff instructing them to test their work-at-home capability. We also checked with the VPN service provider our ability to add additional bandwidth if needed – 60% of the workforce was able to work from home, but looking into the future, we wanted to ensure that if we needed more bandwidth for additional staff we would have it available to us.

The scenario intensified throughout the exercise and we received additional information including:

• flu shot availability in the cafeteria (a red herring);
• a cat 4 Hurricane Mary was off the coast of Bermuda – Important!;
• a fuel tanker accident at the Richmond, VA data center which closed down the facility – major crisis and also on the evening news in VA;
• the CT data center lost all power because a security guard had an adverse reaction to some over-the-counter flu medication, passed out and hit his head on the emergency power shutoff button. Physical security was automatically notified and the police were called to secure the disabled data center,
• Richmond, VA schools were closing at 1 pm due to the hurricane warning.

The IT recovery team notified the Richmond, VA data center recovery service provider that recovery was required; the turnkey arrangement ensured that the data center was up and running within a few hours. The Madison, CT data center had an active-active arrangement with a data center service provider (which the 1998 plan did not identify), so operations were automatically cut over to it once the power went out due to the untimely power shutoff. Prior to the power going out at the CT facility, the Incident commander asked IT to update the employee and franchisee crisis portal with information about how to communicate with Can-Do if the hurricanes hit either location. This activity was not completed because of the power outage in CT.

The power outage resulted in an IT recovery ETA of two days. That information was based on the very old emergency response plan which did not contain the current IT configuration at either data center nor recovery procedures for the current configurations. In reality, there was no loss of IT – even through the power outage due to the prior arrangements made with the DR service providers.

Finance issued a communication requesting that they be notified of any needed hotel and travel arrangements and that all expenses incurred needed to be justified correctly as storm-related or fuel spill-related. Due to the escalating conditions, they subsequently upped the credit limits on all corporate credit cards.

An interesting twist to the exercise was that somehow the FDIC got into the facility and was snooping around asking about Can-Do’s recovery ability. The FDIC person faked her identity to the IT recovery team as it turns out. Though they were smart enough to ask who she was, the team was a bit disjointed and she was given the information she asked for.

Multiple emergency messages were being sent to employees from the various recovery teams throughout the exercise but not everyone was getting the messages. As it turned out, some cell phone towers were down in Richmond, VA due to Hurricane Mary hitting land.

Finally, staff in CT were sent home at 7 pm that day in preparation for the hurricanes hitting.

Hopefully you can sense the problems that arose during this exercise. Some observations:

• The plan needed to be updated ASAP! An old plan is worthless and you waste a lot of time trying to figure out what the current practices are for production processing and recovery of those practices.
• Some exercise participants weren’t aware that they needed to DO something – they sat at the team table and just talked about what they would do in an actual recovery rather than going over to the table of the team with whom they needed to communicate.
• The exercise showed that internal recovery team roles need to be defined in advance so that when an incident occurs, everyone knows their role.
• Communications between the teams was strained at first – no one knew who the coordinators were. Although Can-Do ascribed to NIMS, no crisis command procedures nor crisis communication procedures were part of the plan.
• It was not clear that there was an emergency/mass notification tool available. It took a good 30 minutes for the tool to start being used consistently. AND the first time recovery teams wanted to send a message, they were informed that they needed to have that message approved by the incident command center – also not in the recovery plan and obviously not tested. And rather much a thorn in the side of IT because we thought we should be able to send out our own IT messages – WAKE UP CALL that external communications must be reviewed by the right people internally AND that certain kinds of messages should be set up in advance so that when the interruption occurs, you have much of the messaging in place and approved.
• Releasing recovery status information to someone who turned out to be from the FDIC presented a loophole in the crisis communications process.
• Command center check-in calls with all team coordinators need to be scheduled on a regular basis.
• There was no conference bridge capability for the incident command center to use so everyone had to physically be present at the command center.
• An activity log of actions and tasks and their status needs to be created and updated throughout the incident.
• Finally, a best practice for recovery exercising is to bring in outside observers to identify things during the exercise that you can’t see for yourself.

We had a great event and we all learned a lot about how important it is to have a current plan, clear recovery role assignment, how we personally respond in a crisis and how communications is KEY to recovery success.

U.S again under threat

Published 9 September 2011

New York City and the District of Columbia respond to “specific, credible but unconfirmed” intelligence of an impending attack; information obtained indicates a vehicle-borne bomb; NYPD deploys boats, armored vehicles and a 1,000-member counter-terror force

As the nation prepares to commemorate the tenth anniversary of 9/11, New York City and Washington, D.C. again find themselves responding to an al Qaeda attack threat.

Federal authorities have advised local officials of a “specific, credible but unconfirmed threat” to the cities centered around the commemoration of the World Trade Center and Pentagon attacks.

The intelligence community has developed a “general description”of two or three individuals who may already be in the country. Making the effort more difficult is that the individuals in question have common names.

That intelligence came from the tribal region of Pakistan, from a source acknowledged as having a reliable record by U.S. intelligence officials.

It is believed that the attackers originated their journey in Afghanistan, with a possible third-country waypoint. That third country may have been Iran.

In the language of counter-terror officials, “specific” means that there is information of the type of attack that may occur. In this case, the information indicated that a vehicle-borne explosive device, a car- or truck-bomb, was the chosen method.

Last night, official focus was on two missing rental trucks, from different rental agencies in the Kansas City, Kansas area. They were later found and determined to be unconnected with the present threat.

“Credible” is used to indicate that the source of the information is believable, comes from a reliable, knowledgeable source. U.S. signals intelligence has been listening in on the communications of one particular al Qaeda source in Pakistan, from whom officials have gleaned confirmed information in the past.

Also supporting the credibility of the intelligence is an increase in “chatter” on the communication channels that are known to be used by al Qaeda operatives.

In the trove of documents gathered from Osama bin Laden’s compound in Abottabad, Pakistan, during the raid that killed him, bin Laden showed a predilection for attacking the United States on significant dates and anniversaries, such as the upcoming 9/11 commemorations.

What has not yet been uncovered is the type of corroboration that would provide confirmation indicating that the plot is active and in progress.

New York City wasted no time in responding to the threat notification.

All bridges and tunnels entering Manhattan have been staffed with additional police and national guard personnel. Cars and trucks entering the city are being searched. Additionally, there have been checkpoints set up at various locations in Manhattan, such as Times Square and Lower Manhattan, approaching the financial district.

Key rail and subway stations operated by the Metropolitan Transit Authority, Port Authority of NY and NJ and NJ Transit have been staffed with additional officers accompanied by national guard troops, watching traffic outside stations and randomly searching backpacks and baggage inside.

New York City’s response has been thorough. Besides police officers at the bridges, tunnels and rail stations, the city has deployed radiation-detecting boats, cameras have been placed throughout midtown and lower Manhattan. If required, the NYPD has a small, unmanned submersible craft available to search the hulls of ships and boats.

Also deployed or on standby, is an “army” of 1,000 anti-terror officers, armored vehicles and weapons and EOD (explosive ordinance disposal) specialists.

According to FEMA  “this new resource educates individuals and families about how using modern-day technology can help them prepare, adapt and recover from disruptions brought on by emergencies or disasters. Get Tech Ready provides Americans with tips on how to use technological resources before, during and after a crisis to communicate with loved ones and manage your financial affairs. Preparedness tips on the website include:

• Learn how to send updates via text and internet from your mobile phone to your contacts and social channels in case voice communications are not available;
• Store your important documents such as personal and financial records in the cloud or on a secure and remote area or flash or jump drive that you can keep readily available so they can be accessed from anywhere; and
• Create an Emergency Information Document using the Ready.gov Family Emergency Plan template in Google Docs or by downloading the Ready Family Emergency Plan to record your emergency plans.”

Check it out and let us know what you think.

Also, it seems that emergency/mass notification services (EMNS) were used extensively to alert the population of storm status: NYC through NotifyNYC and NYC-OEM sent regular pre- and post- alerts regarding the event, I received voicemails or emails from my local CT town management, Connecticut Light & Power, and JPMorganChase alerting me about disaster preparedness status and steps to take if I needed assistance.

Another new feature was the use of texting: If one texted the name “Irene” to 501-01, National Grid would text regular updates on electrical power restoration status to your cell phone.    This feature definitely was not around back in the days of Hurricane Gloria (1985) or Bob (1991) and was quite useful since over 500K National Grid customers lost power.

Also, going to wifi hot spots at venues like Starbucks and McDonald’s is certainly a new capability. How many of you used one of these options?

And, on August 26, 2011 FEMA launched its first-ever smartphone application and text messaging updates. Available right now only on the Android smartphone, Blackberry and iPhone support will be coming in a few weeks.

What were your experiences if you were in an impacted area?

Roberta Witty and John Morency

Key Findings

• The earthquake and tsunami that struck the Tohoku district in March, and the power plant failures and other infrastructure problems that followed, continue to disrupt communications, transportation and other infrastructure.
• The Japanese government and Tepco have implemented a plan for rolling electrical blackouts across Tepco’s coverage area, designed to reduce power usage and avoid total power failures.
• These blackouts present serious challenges for Japanese enterprises, particularly in maintaining the operational integrity of their data centers and offering alternative system access to remote workers.

Tepco has said it will not carry out its planned rolling blackouts this summer, but electrical supply continues to present challenges for Japanese enterprises. Gartner has developed a set of best   practices for various scenarios and affected parties for IT organizations in Japan and worldwide. The appropriate response to the rolling blackout depends heavily on whether or not the enterprise’s data center has its own dedicated backup power generator.

Read more about the best practices – if you are an organization impacted by the earthquake/tsunami or not – in the full report by my colleagues  Masahiko Ishibashi, Eiichi Matsubara, Nagayoshi Nakano and Katsuo Hori.  Being a Gartner customer may be required.

The findings in this research are based on a joint Gartner and Business Continuity Institute (BCI) survey of business continuity management (BCM) consulting firms conducted during the first quarter of 2011. The survey objective was to better understand the breadth and depth of BCM service offerings. It was sent by BCI to its self-identified consultant membership and by Gartner to members on its BCM consultancy list that opted to participate. The survey closed on 4 March 2011.

Key Findings

• Many firms say they cover many BCM disciplines, but terminology varies across industries and countries, and so, misunderstandings are common.
• There is an increased need for consultants with specific skill sets:
  • Strategic program development
  • Tactical program improvements
  • Pragmatic, situation-based expertise
• A strong BCM program cannot be run by consultants alone. Therefore, BCM expertise must be brought in-house to ensure its continuing success.
• Due to the 2008 global financial crisis (GFC), consultant ranks have risen due to the layoffs of BCM professionals, and many of these people have taken jobs as consultants while waiting for a full-time BCM practitioner position.
• Fifty percent of BCM consultancies are small, with one to four full-time consultants onboard.
• Organization certification support is low — 13% of firms surveyed have BS 25999 Lead Auditor certification — in alignment with the existing low level of organizations that have such certification.
• Ninety-two percent of BCM consulting engagements are for planning services.
• Only one-quarter (24%) of BCM consulting firms offer a guarantee for services rendered.

Read the full report here: BCM Consultancy Survey, 2011. You may need to be a Gartner client to access the report.

Granted, excessive snow on roofs is a rather uncommon event (even in New England), and given the number of unique crises in 2010 and 2011 – volcanic ash and a government countrywide shut down of Internet and cell phone access to name just two, the risks facing BCM managers is growing and they are getting lots of great experience in treading new tracks (snow management tracks in this case). So the question is why did the Iron Mountain situation occur given the weeks of news reports of such events taking place?

The underlying reasons for this collapse have not been made public by Iron Mountain, but I can tell you that BCM service providers are a critical group of vendors that must be included in your supply chain availability risk management program. Not just from the planning perspective, but from the crisis management perspective as well. How many firms stood up their crisis command center and pro-actively reached out to their vendors since December/2010 to understand how they were handling the excessive snow and how secure their assets are and will be if the situation gets worse? Not many I’m sure. The 2009 H1N1 crisis was one situation where some firms did do such outreach.

Customers of every records management vendor, or any vendor that houses your organization’s assets, should understand in detail what are the vendor’s guarantees regarding the safeguarding of assets in unique circumstances such as excessive snow, and understand what recourse you have if they are damaged or if the situation is too risky for them to remain in their current facility. You should also require an annual assessment report of the vendor’s own BCM plan for the facility in which your assets are stored.

If you know of other such events for BCM service providers, if your firm has stood up their crisis command center this winter or have your own experience in these situations, we’d love to hear from you.

I’m wondering what tools your firm is using during such events? They have to be hosted by the vendor since your data center is down. What about Yammer, SocialCast, SocialText Signals, Sametime/LotusLive from IBM, Lync/BPOS from Microsoft, Google Apps? Is anyone using them or another tool for emergency purposes, and if so, how?

Business Continuity Management Governance Defined, 2010

Without a governance framework in place, BCM programs will not progress as needed in the desired time frame. Use Gartner’s BCM governance framework to establish governance oversight according to your organization’s business model and availability needs.

Toolkit: BCM Governance and Implementation Responsibility Decision Matrix, 2010

Without a governance structure in place, business continuity management programs will not progress or succeed in the time frame desired. Use the BCM Governance and Implementation Responsibility Decision Matrix to document the governance oversight according to your organization’s business model.

Toolkit: Business Continuity Management Charter Best Practices and Template

A BCM charter is one of the most-effective tools for establishing and communicating effective preparedness, recovery and resiliency practices. Make the Gartner best-practice-based charter the foundation of an enterprisewide BCM program.

Toolkit: Business Continuity Management Policy Template

A business continuity management policy is an important component of the operational model for BCM governance. Combined with a BCM charter, BCM governance responsibility and implementation matrix, and the BCM activity cycle, it completes the Gartner BCM governance framework.

Not only is it National Preparedness Month, it’s prime time hurricane season in the northern hemisphere.  Watch for alerts about the status of storms at NOAH’s National Hurricane Center.

The American Red Cross has excellent advice about preparing your home and family for a disaster.

Ready.gov – a FEMA sponsored web site – is another great resource for family disaster preparedness.

Gartner has a cornucopia of research and advice including templates, vendor analysis, best practices, sample RFPs and more for organizations that are developing, maintaining and maturing their BCM programs. Read our note:  Research Roundup: Business Continuity Management and IT Disaster Recovery Management, 2Q10 to see the breadth and depth of our coverage on all things BCM.

And watch for a new BCM tool coming from Gartner in the coming weeks. Can’t say too much about it now, but organizations are so primed for it!

So take this time to promote, assess, enhance and EXERCISE your recovery plans and procedures. If you don’t, you are missing an opportunity to get management’s attention and commitment in protecting its workforce, customers and partners from disaster events.

I’m going to check and update my Go Bag today. I found a great First Aid kit through www.melaleuca.com. I’ve already used it once at a local retail outlet.  Who knew it would come in handy so quickly after purchase!

One last thing: have you given a friend or family member a list of your user IDs and passwords for personal web sites such as healthcare, insurance, online banking, investments, email, social media applications and so forth – just in case – well you know why.