DOJ’s 2016 bank DDoS indictment described lost customer access, not stolen data; a 2021 rule named that harm

The March 24, 2016 indictment alleged a DDoS campaign that cut bank customers off from accounts without taking account data. Bank regulators later gave an extended outage of that kind as an example of a notification incident. Customer access is a resilience duty in itself.

Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.

Two-minute summaryThe dek and the first sentence of each section, quoted from the article

The March 24, 2016 indictment alleged a DDoS campaign that cut bank customers off from accounts without taking account data. Bank regulators later gave an extended outage of that kind as an example of a notification incident. Customer access is a resilience duty in itself.

  1. The week The release said the grand jury indicted seven Iranian individuals employed by two Iran-based computer companies. [1] It said the companies performed work on behalf of the Iranian government, including the Islamic Revolutionary Guard Corps. [1] We do not name the defendants.
  2. The harm: access taken, data left The release described three harms: disabled bank websites, customers unable to reach their accounts online, and tens of millions of dollars in remediation costs. [1] It described no loss of customer data. [1] The asset attacked was the bank’s ability to serve customers through a digital channel.
  3. Capacity: who absorbs the traffic The release said victim servers were hit with as much as 140 gigabits of data per second on certain days. [1] It said the botnets consisted of thousands of compromised systems owned by unwitting third parties. [1] An attack at that scale arrives through the internet providers upstream of the bank.
  4. Communication: customers, peers, providers and regulators The FFIEC said response plans should include appropriate communication strategies with customers concerning the safety of their accounts. [3] During a DDoS attack, customers see a slow or unavailable site.
  5. Attribution: what a public indictment changes Public charges of this kind name individuals and describe methods.
  6. How the duty differs by institution A large bank runs its own digital channels and may have direct relationships with several internet providers.
  7. Objections to the argument, in order No account data was taken, according to the release. [1] Customers still lost access, and banks spent tens of millions of dollars on remediation, as the release described. [1] The 2021 rule treats lost customer access of that kind as a possible notification incident. [4] The rule’s authors chose it as an example of a notification incident.
  8. Checks a bank or credit union could run Assign one executive to own the availability of each customer channel: online banking, mobile, payments portals.
  9. With hindsight Bank regulators later wrote a reporting duty that reaches this kind of harm.

The week

On Thursday, March 24, 2016, the Department of Justice announced the unsealing of an indictment returned by a grand jury in the Southern District of New York. [1] We read the release in an Internet Archive copy captured on April 2, 2016, because the live Justice Department page now answers automated requests with a browser check. [1]

The release said the grand jury indicted seven Iranian individuals employed by two Iran-based computer companies. [1] It said the companies performed work on behalf of the Iranian government, including the Islamic Revolutionary Guard Corps. [1] We do not name the defendants. The charges are allegations, and the release itself said an indictment is merely an accusation and all defendants are presumed innocent unless proven guilty. [1]

According to the release, the charges related to an extensive campaign of over 176 days of distributed denial of service (DDoS) attacks. [1] It said the attacks targeted 46 victims, primarily in the U.S. financial sector, between late 2011 and mid-2013. [1]

The release described the harm alleged. It said the attacks disabled victim bank websites, prevented customers from accessing their accounts online, and collectively cost the victims tens of millions of dollars in remediation costs. [1] It also said the attacks did not affect or result in the theft of customer account data. [1]

The release gave a timeline drawn from the indictment. It said the campaign began in approximately December 2011, and attacks occurred only sporadically until September 2012. [1] From then, it said, they escalated to a near-weekly basis, between Tuesdays and Thursdays during normal business hours in the United States. [1]

It described the scale. On certain days, the release said, victim servers were hit with as much as 140 gigabits of data per second. [1] On those days, it said, hundreds of thousands of customers were cut off from online access to their bank accounts. [1]

The release described the method. It said each group built and maintained its own botnet of thousands of compromised computer systems owned by unwitting third parties. [1] Orders sent to the botnets directed malicious traffic at servers used to operate victim financial institutions’ websites, overwhelming them. [1]

It also described what law enforcement had done since. The release said the FBI provided updated information about infected systems through approximately 20 FBI Liaison Alert System (FLASH) messages. [1] It said the FBI conducted outreach to internet service providers hosting infected systems. [1] It said that, through these efforts, over 95 percent of the known part of the botnets had been remediated. [1]

A separate count concerned a dam. The release said one defendant was charged with obtaining unauthorized access to the control systems of the Bowman Dam in Rye, New York, between August 28 and September 18, 2013. [1] It said the sluice gate had been manually disconnected for maintenance at the time, so remote operation was not possible. [1]

For banks, the attacks were not new information in March 2016. On April 3, 2014, the OCC had issued Bulletin 2014-14, attaching a joint statement by the members of the Federal Financial Institutions Examination Council (FFIEC). [2] The bulletin said the statement notified institutions of the risks of continued DDoS attacks and the steps institutions were expected to take. [2]

The FFIEC statement said that in the latter half of 2012, an increased number of DDoS attacks were launched against financial institutions by politically motivated groups. [3] It said the attacks caused slow website response times, intermittently prevented customers from accessing public websites, and adversely affected back-office operations. [3] In other cases, it said, DDoS attacks served as a diversion for criminals attempting fraudulent wire or ACH transfers. [3]

The FFIEC members said they expected each institution to address DDoS readiness as part of its information security and incident response plans. [3] The OCC bulletin added a note for community banks: they should ensure that in-house IT units or service providers were taking appropriate action. [2]

The March 2016 indictment is remembered for its attribution: the Justice Department publicly charged individuals it tied to a foreign government’s work. [1] For bank technology and risk leaders, the more useful part of the release was its description of the harm. Customers were cut off from their accounts; according to the release, account data was not taken. [1]

Our argument is that the case marks availability of customer access as a resilience duty in its own right, separate from data protection. The FFIEC had pointed that way in 2014, when it described slow sites, customers intermittently prevented from reaching public websites, and back-office effects. [3] In 2021, the federal banking agencies gave a large-scale DDoS attack that disrupts customer account access for an extended period of time as an example of a notification incident. [4]

We take the duty in four parts:

  1. The harm: what a DDoS attack takes from a bank, and why it differs from a data breach.
  2. Capacity: who absorbs attack traffic, and what the bank must arrange before an attack.
  3. Communication: what customers, peers, providers and regulators need to hear, and when.
  4. Attribution: what a public indictment changes for a bank, and what it leaves unchanged.

The parts do not overlap. The first concerns the asset at risk, the second the technical defense, the third the information flows, and the fourth the role of law enforcement. After them we look at how the duty differs by institution, take the objections in order, and end with checks a bank or credit union could run.

The harm: access taken, data left

What was lost

The release described three harms: disabled bank websites, customers unable to reach their accounts online, and tens of millions of dollars in remediation costs. [1] It described no loss of customer data. [1] The asset attacked was the bank’s ability to serve customers through a digital channel.

The FFIEC’s 2014 statement listed similar effects from attacks it said began in the latter half of 2012: slow response times, customers intermittently unable to reach public websites, and adverse effects on back-office operations. [3] It named the risks as operational and reputational, and added fraud, liquidity and capital risks where an attack was coupled with attempted fraud. [3]

Duration as the measure

The release described a campaign with a rhythm: sporadic attacks at first, then near-weekly attacks between Tuesdays and Thursdays during business hours. [1] That pattern means the harm is cumulative. Each attack may be survivable, while the series wears down staff, budgets and customer patience.

A resilience plan written for a single event may not fit a campaign. It should say how the bank sustains its response over repeated attacks: staff rotation, provider capacity, and how customer messages change when the problem recurs. Planning for a series is a different exercise from planning for a day.

Why it differs from a breach

A data breach is measured by what was taken. A DDoS attack is measured by what could not be done, and for how long. That difference changes the controls, the metrics and the owners. Encryption and access control protect data. Capacity, routing and failover protect availability.

Customer data rules and breach notification laws point a security program toward confidentiality. An availability attack tests a different part of the program, one that sits between the security team, the network team, the digital banking team and the vendors that host the channel.

The customer’s view

For customers, the harm was concrete. The release said hundreds of thousands of customers were cut off from online access to their bank accounts on certain days. [1] A customer who cannot see a balance or move money does not distinguish between an attack and an outage.

That is why we treat customer access as a service with its own owner. Someone must be accountable for how long customers can be without it, what alternatives they have, and what they are told. In a breach, those questions arise after the fact. In a DDoS attack, they arise while the attack is under way.

Every customer channel

The release spoke of bank websites and online account access. [1] Today a bank’s customers also reach it through mobile apps, programming interfaces used by aggregators and partners, and payment services. Each depends on internet-facing infrastructure that a DDoS attack can target.

A bank’s inventory of customer access should list all of them, with the infrastructure each relies on. A mobile app that calls the same back-end servers as the website shares the website’s exposure. An interface used by partners may carry customer traffic the bank does not see as its own channel.

Back-office effects

The FFIEC said the attacks adversely affected back-office operations. [3] Shared infrastructure explains how. Internet links, firewalls and domain name services can serve public websites and internal functions at once. Traffic aimed at the website can slow everything that shares the path.

A bank should know which internal functions share infrastructure with its public channels. Payment operations, treasury connections and remote staff access are candidates. If they share a path, a DDoS attack on the website is also an attack on them.

The diversion risk

The FFIEC added a warning that the release did not address. It said that in some cases DDoS attacks served as a diversionary tactic by criminals trying to commit fraud with stolen customer or employee credentials, through wire or ACH transfers. [3]

That makes the harm two-sided. While staff respond to the visible outage, the fraud team must watch the less visible payment flows. A response plan that pulls fraud analysts into the outage response leaves the payment side thinner at the moment it most needs attention.

Measuring availability

A bank cannot manage a harm it does not measure. Useful measures for customer access include login success rates, transaction completion rates, response times by channel, and the number of customers affected. They should be measured from outside the bank’s network, where customers are, as well as inside it.

Those measures also feed the notification decision we discuss below. Without them, a bank has to decide whether access was materially disrupted from impressions. With them, it can point to numbers it collected at the time.

Capacity: who absorbs the traffic

Scale beyond one bank

The release said victim servers were hit with as much as 140 gigabits of data per second on certain days. [1] It said the botnets consisted of thousands of compromised systems owned by unwitting third parties. [1] An attack at that scale arrives through the internet providers upstream of the bank.

No bank can buy enough bandwidth to absorb every attack. The practical defense is to filter traffic before it reaches the bank: at the internet provider, at a mitigation service, or at a content delivery network in front of the public site. Each of those is a third party.

What the FFIEC expected

The FFIEC statement listed six steps it expected institutions to take, as appropriate. [3] In order, they were: to maintain a program to assess information security risk, including threats to external websites and online accounts; and to monitor internet traffic to the website to detect attacks. [3]

The third was to activate incident response plans and notify service providers, including internet service providers, if an attack was suspected. [3] The fourth was to ensure sufficient staffing for the attack’s duration and consider pre-contracted third-party servicers to help manage traffic flow. [3]

The fifth was to consider sharing information with organizations such as the Financial Services Information Sharing and Analysis Center and law enforcement. [3] The sixth was to evaluate gaps in the response after attacks and adjust controls. [3]

Pre-contracted help

The fourth step depends heavily on preparation. The FFIEC said institutions should consider hiring pre-contracted third-party servicers, and identify how their internet service provider could assist. [3] A contract signed during an attack is slower and costlier than one signed before.

A mitigation contract should say what traffic the provider will filter, how fast it will act once called, how the bank calls it, and what it costs during a long campaign. The release described attacks on a near-weekly basis from September 2012. [1] A contract priced per incident may behave very differently under that pattern than one priced by subscription.

The community bank case

The OCC’s bulletin addressed community banks directly. It said they should ensure that their in-house IT units or service providers were taking appropriate action. [2] For a community bank whose online banking is hosted by a provider, capacity is a contract question.

Such a bank should know whether its provider has mitigation in place, whether it covers the bank’s domain names, and what the provider will do if several of its client banks are attacked at once. Shared hosting means shared exposure. One bank’s attack can slow another bank’s customers.

Domain names and the edge

Attack traffic can target more than web servers. CISA’s 2024 guide listed difficulty reaching other network services, such as DNS servers or firewalls, as a possible sign of an attack. [6] If the service that resolves a bank’s domain names fails, customers cannot reach a healthy website.

A bank should know who provides its authoritative domain name service, whether that provider has its own DDoS protection, and whether a secondary provider exists. Those questions are easy to skip, because domain services may go years without review. They belong in the same review as the website’s mitigation.

Cost over a campaign

The release said the victims collectively spent tens of millions of dollars on remediation. [1] That figure covers many institutions over a long period, and we do not divide it. It shows that the cost of defense falls on the victim, whatever happens to the attackers.

Budgeting for DDoS readiness means paying for capacity that sits idle most of the time. The case for that spending is the cost of lost access: customer attrition, call center load, staff time and, since 2022 for covered banks, a regulatory notice. A board that understands that trade-off can approve the spend knowingly.

Testing

A mitigation arrangement that has never been switched on is an assumption. Banks can test routing changes, provider escalation and failover without a live attack, in planned windows. The test should include the people: who calls the provider, who approves a routing change, and who confirms customers can log in again.

CISA, the FBI and MS-ISAC later listed proactive steps in their 2024 guide, written for government entities, including a DDoS mitigation service, bandwidth capacity planning, load balancing, and redundancy and failover. [6] The list is a reasonable checklist for a bank’s own test plan, even though banks were not its audience. [6]

Collective cleanup

The release described a part of the defense no bank controlled. It said the FBI provided information about infected systems in approximately 20 FLASH messages and conducted outreach to internet service providers. [1] It said over 95 percent of the known part of the botnets had been remediated through those efforts. [1]

Cleanup of infected computers reduces the traffic future attacks can generate. It depends on providers and owners of those computers acting on information. Banks benefit from it without paying for it, which is one reason information sharing is part of the FFIEC’s list rather than an optional extra.

Communication: customers, peers, providers and regulators

Customers

The FFIEC said response plans should include appropriate communication strategies with customers concerning the safety of their accounts. [3] During a DDoS attack, customers see a slow or unavailable site. They do not know whether their money is safe.

A bank that can say quickly and accurately that the problem is access, not account data, reduces call volume and anxiety. It can say so only if it knows. That requires a fast way to confirm that the attack is a DDoS attack and not cover for something else, which loops back to the diversion risk the FFIEC described.

What not to say

Early in an attack, a bank may not know whether data is safe. A message that says no data was affected, sent before that is known, can become a problem if it proves wrong. The safer message describes what the bank knows: customers cannot reach the service, the bank is working on it, and other channels are open.

The release’s statement that account data was not taken came at the end of an investigation, not during the attacks. [1] A bank in the middle of an attack has no such record. Its messages should match what it can confirm at the time, and change as it learns more.

Alternative channels

Customers cut off online will call, visit branches or use ATMs. Those channels need warning and capacity. A communication plan should tell customers which channels work, and should tell branch and call center staff what to say.

Peers

The FFIEC encouraged institutions to consider sharing information with the Financial Services Information Sharing and Analysis Center and law enforcement, because attacks can change rapidly. [3] The release showed the other side of that flow: the FBI sending FLASH messages about infected systems. [1]

Sharing works in both directions only if the bank has a named person who can send and receive under agreed rules. That person should be part of the incident response plan, with authority to share technical indicators without waiting for executive sign-off on each message.

Providers

The FFIEC told institutions to notify service providers, including internet service providers, when an attack was suspected. [3] The 2021 federal banking rule later added a duty in the other direction for bank service providers. [4]

Under that rule, a bank service provider must notify each affected banking organization customer as soon as possible after it determines it has had a computer-security incident. [4] The duty applies where the incident has caused, or is reasonably likely to cause, a material service disruption or degradation for four or more hours. [4]

Staff

Staff need to know what is happening before customers ask them. Branch, call center and relationship staff should receive a short internal notice early in an attack, with the approved customer message and the alternatives available. If staff learn of the outage from customers, the bank’s message is already behind.

Internal communication also has to reach the people who run payments and treasury. They may see effects on back-office systems that share infrastructure, as the FFIEC described. [3] Their reports help the response team judge whether the attack has spread beyond the website.

The board

A board does not need minute-by-minute updates during an attack. It does need a report afterward that says what customers lost, for how long, what it cost, whether a regulator was notified, and what will change. That report turns an incident into a governance record.

Regulators

The 2021 federal banking rule set a deadline for telling a bank’s primary federal regulator about such an incident. For banking organizations it covers, it requires notice as soon as possible and no later than 36 hours after the banking organization determines that a notification incident has occurred. [4]

The rule’s preamble gave examples of notification incidents. One was a large-scale distributed denial of service attack that disrupts customer account access for an extended period of time. [4] In our reading, that example describes the harm the 2016 release alleged, in a regulator’s words.

The determination

The 36-hour clock in the 2021 rule starts at the bank’s determination, not at the start of the attack. [4] That places weight on how quickly and how well the bank decides. A bank that has defined, in advance, what level of lost access counts as material can decide faster and defend its decision later.

Defining that level is a business decision with a security input. It should consider the number of customers affected, the duration, the channels lost and the availability of alternatives. The definition should be written, approved and tested in an exercise before it is needed.

Attribution: what a public indictment changes

What the charges said

The release tied the defendants to companies it said performed work for the Iranian government, including the Islamic Revolutionary Guard Corps. [1] It described the charges as computer hacking conspiracy for the DDoS campaign and a separate count for the dam intrusion. [1]

Public charges of this kind name individuals and describe methods. They also signal that the government regards the campaign as a national security matter. For a bank, that signal has some value: it confirms that the attacks came from a capable, persistent actor and were not random.

What the charges did not change

The indictment did not change any bank’s obligations, its capacity or its contracts. The release placed the attacks between late 2011 and mid-2013. [1] Defending against the next campaign depended on what banks had built in the meantime.

Attribution also does not tell a bank who will attack next. It describes one campaign in the past. The FFIEC’s steps, written before the indictment, did not depend on knowing the attacker, and that independence is what made them useful.

Evidence and cooperation

The release said the case was investigated by multiple FBI field offices. [1] An investigation of this kind can draw on victims’ logs and records. A bank that keeps traffic data, timestamps and provider reports during an attack makes later investigation possible.

CISA’s 2024 guide listed gathering evidence among its response steps, including timestamps, IP addresses, packet captures and logs, as useful for reporting to law enforcement or later analysis. [6] A bank’s incident plan should say who preserves that evidence and for how long, and counsel should review the retention rules that apply.

Information from government

The release described one concrete benefit banks received from the investigation: information about infected systems, shared in FLASH messages. [1] That information let providers and owners clean up machines that would otherwise carry the next attack.

To use such information, a bank or its provider must be able to receive it, read it and act on it. That means a named contact, a process for turning indicators into blocking rules, and a way to share what the bank sees in return. The value of government information depends on that intake.

Attribution and customers

A public indictment can change how customers and the press talk about past outages. A bank named or recognizable as a victim may face questions about events from years earlier. Its communications team should be ready to say what happened, what has changed since, and what it cannot discuss because of an investigation.

That readiness depends on records. A bank that kept a clear account of past attacks, its response and its improvements can answer those questions accurately. One that did not will find itself reconstructing events from memory.

The dam allegation

The dam count belongs to another sector, but it carries a point for banks. The release said the intruder could obtain information about water levels, temperature and the sluice gate, and that remote operation was not possible because the gate had been manually disconnected for maintenance. [1]

In our reading, the detail shows that the same actors described in the release were interested in more than availability attacks. A bank that reads the case as only a DDoS story may miss that broader interest. Threat assessments should consider the actor’s full alleged range, as well as the technique used against banks.

Sanctions and payments

An indictment of foreign individuals can be followed by other government actions. Banks should route any such development through their sanctions compliance function, which has its own obligations. We did not research later actions tied to these defendants and make no claim about them.

How the duty differs by institution

A large bank

A large bank runs its own digital channels and may have direct relationships with several internet providers. Its challenge is coordination across many teams and providers during an attack. It is covered by the 2021 federal banking rule if it falls within the rule’s definition of a banking organization. [4] Counsel should confirm coverage for each legal entity.

A community bank

A community bank may rely on providers for hosting and for security operations. The OCC told community banks to ensure their in-house IT units or service providers were acting on the DDoS risk. [2] The 2021 rule’s service provider notice duty gives such a bank a contractual and regulatory hook to hear about provider incidents. [4]

A credit union

Federally insured credit unions are covered by NCUA’s rule, published March 1, 2023 and effective September 1, 2023. [5] It requires notice to NCUA as soon as possible and no later than 72 hours after the credit union reasonably believes it has experienced a reportable cyber incident. [5] NCUA said a DDoS attack that disrupts member account access will be reportable under one prong of its definition. [5]

NCUA’s rule differs from the banking agencies’ in its clock and its trigger: 72 hours from reasonable belief, against 36 hours from determination. [4] [5] A credit union should not assume a bank-oriented incident playbook fits its rule without adjustment.

A bank service provider

A provider hosting online banking for many institutions faces the attack on behalf of all of them. Under the 2021 rule, it must notify each affected banking organization customer as soon as possible once it determines that a qualifying incident has occurred. [4] The rule’s threshold is a material disruption or degradation, actual or reasonably likely, lasting four or more hours. [4] Its contracts should already say how that notice is sent and to whom.

A broker-dealer or investment adviser

Securities firms are not banking organizations under the 2021 rule, and their regulators set their own expectations. The availability lesson applies to their client portals and trading interfaces all the same. They should take the notification question to counsel under their own rules.

A fintech partner of a bank

A fintech that delivers a bank’s products through its own app is part of the bank’s customer access, even if the bank does not host it. Its availability affects the bank’s customers. The bank’s third-party risk program should ask the fintech the same capacity, contract and communication questions the bank asks itself, and agree how each will tell the other about an attack.

A state or local government

The 2024 CISA, FBI and MS-ISAC guide was written for federal, state, local, tribal and territorial government entities. [6] Public portals for taxes, benefits and payments face the same availability risk. The guide’s preparation and response lists give such agencies a starting point for the same four parts we describe. [6]

Objections to the argument, in order

1. A DDoS attack is a nuisance, not a breach

No account data was taken, according to the release. [1] Customers still lost access, and banks spent tens of millions of dollars on remediation, as the release described. [1] The 2021 rule treats lost customer access of that kind as a possible notification incident. [4] The rule’s authors chose it as an example of a notification incident.

2. Attribution does not help a bank defend itself

We largely agree, and the fourth part says so. Attribution confirms the seriousness of a campaign and supports law enforcement. The defense rests on capacity and communication built beforehand.

3. Mitigation is the internet provider’s job

Providers carry much of the technical work. The FFIEC still expected each institution to address DDoS readiness in its own plans, notify providers and arrange third-party help. [3] The bank remains accountable for its service to customers, whoever filters the traffic.

4. Notification rules add burden for short outages

The 2021 rule applies to incidents that materially disrupt or degrade, or are reasonably likely to, certain bank operations or services. [4] In our reading, a brief slowdown that does not reach that level is not a notification incident under the rule’s definitions. Each bank should apply the definitions with counsel, but the rule is not written for every outage.

5. The 2012 attacks are old; today’s are different

Techniques change. CISA’s 2024 guide sorts DDoS techniques into volumetric, protocol and application types. [6] The duty we describe does not depend on a technique. It depends on the bank knowing what customer access it must protect, who absorbs attack traffic, and who must be told.

6. The charges are allegations, so drawing lessons is premature

The charges are allegations, and the release said so. [1] Our lessons do not depend on any defendant’s guilt. They rest on the harms the FFIEC described in 2014 and the rules regulators wrote later. [3] [4] [5] The release adds a description of scale and method that is consistent with those documents.

7. Cloud hosting removes the problem

Hosting public channels with a large cloud or content provider can add capacity and filtering that a bank could not build alone. It moves the capacity question to a contract, as with any provider. The bank still needs to know what protection the contract includes, how the provider reports attacks, and what happens if the provider itself is disrupted.

Concentration adds a further question. If many banks rely on the same provider, an attack on or failure at that provider can affect them together. The agencies said they kept the service provider notice duty because of the increasingly significant role that bank service providers play in the banking industry. [4]

Checks a bank or credit union could run

Name the owner of customer access

Assign one executive to own the availability of each customer channel: online banking, mobile, payments portals. Give that owner the metrics, the provider relationships and the authority to act during an attack.

Map the shared paths

Identify which internal functions share internet links, firewalls or domain services with public channels. Decide whether to separate them, or accept the shared risk in writing.

Read the mitigation contracts

For each provider in front of a customer channel, confirm what mitigation it provides, how fast, how it is called and at what cost during a long campaign. Confirm the contract covers every domain name customers use.

Test the switch

Run a planned exercise that activates mitigation and confirms customers can log in during it. Include the people who call providers and approve changes. Record the times each step took.

Define material disruption

Write down, with counsel, what level of lost customer access your institution will treat as a notification incident under the 2021 rule or NCUA’s rule. [4] [5] Base it on customers affected, duration and alternatives. Practice the determination in a tabletop exercise.

Watch payments during attacks

Add a step to the DDoS playbook that raises fraud monitoring on wire and ACH transfers for the duration, as the FFIEC’s warning about diversion suggests. [3] Keep fraud staff on payment monitoring rather than pulling them into the outage response.

Prepare customer messages

Draft messages that explain an access problem and say what customers can do instead. Prepare a separate message for the case where account safety is in question. The FFIEC asked for communication strategies on the safety of accounts. [3]

Measure from outside

Set up monitoring that tests login and transactions from outside your network, from several locations. Use it to measure customer impact during attacks and exercises. Keep the data with the incident record.

Report to the board

After each significant attack or exercise, give the board a short report on customer impact, duration, cost, notifications made and changes planned. Track the changes to completion.

Plan for a campaign

Extend the DDoS playbook beyond a single day. Set staff rotations, provider cost limits and message updates for attacks that recur over weeks. Exercise a scenario with repeated attacks, not a single one.

Agree notices with partners

For each service provider and fintech partner that touches customer access, agree in writing who tells whom about an attack, how fast and through which contact. Check that the arrangement meets the 2021 rule’s service provider duty where it applies. [4]

Keep the evidence

Name who preserves traffic data, provider reports and timestamps during an attack, and for how long. CISA’s 2024 guide lists such evidence as useful for reporting to law enforcement and later analysis. [6] Have counsel confirm the retention period and any privacy limits on what is kept.

Review after every attack

After each attack or exercise, record gaps and adjust controls, as the FFIEC’s sixth step expected. [3] Share what you learned with your information sharing organization, under its rules.

With hindsight

Bank regulators later wrote a reporting duty that reaches this kind of harm. On November 23, 2021, the OCC, the Federal Reserve Board and the FDIC published a final rule on computer-security incident notification. [4] It requires a banking organization to notify its primary federal regulator as soon as possible and no later than 36 hours after determining that a notification incident has occurred. [4]

The rule took effect on April 1, 2022, with a compliance date of May 1, 2022. [4] Its preamble listed, among examples of notification incidents, a large-scale distributed denial of service attack that disrupts customer account access for an extended period of time. [4] It also required bank service providers to notify affected banking organization customers of incidents causing, or reasonably likely to cause, material disruption for four or more hours. [4]

The agencies explained the purpose in terms of early awareness. They said the notice would help them react to emerging threats before they become systemic. [4] In our reading, that reasoning fits a campaign of the kind the 2016 release described, which it said targeted 46 victims, primarily in one sector. [1] [4]

NCUA followed for credit unions. Its final rule was published March 1, 2023 and took effect September 1, 2023. [5] It requires a federally insured credit union to report a reportable cyber incident as soon as possible and no later than 72 hours after it reasonably believes one has occurred. [5] NCUA said a DDoS attack that disrupts member account access will be reportable under one prong of its definition. [5]

Federal guidance on DDoS also matured outside banking. On March 21, 2024, CISA, the FBI and MS-ISAC published version 2.0 of a joint guide, first issued in October 2022, written for government entities. [6] It sorted techniques into volumetric, protocol and application types, and listed proactive steps, signs of an attack and response steps. [6]

Its steps include a risk assessment, network monitoring, an incident response plan, a mitigation service, notifying service providers and a post-incident analysis. [6] Each has a counterpart in the FFIEC’s 2014 list. [3] In our reading, the core of DDoS readiness did not change between 2014 and 2024. What changed was that regulators began to require notice when customer access is lost.

The rule also gave banks time. The agencies said they set the compliance date after commenters asked for more time to implement the rule. [4] Some commenters described perceived challenges in renegotiating service provider contracts, and one said community banks may hold little power in those negotiations. [4]

This post does not report the later status of the criminal case. The live Justice Department release could not be opened by our tools on November 16, 2026, and we did not rely on any other record of the case. [1] The release itself said the defendants were presumed innocent unless proven guilty in a court of law. [1]

What held up from March 2016 was the description of harm. Lost customer access, without stolen data, became a named example in a federal banking rule and a named reportable event in NCUA’s rule. [4] [5] What did not hold up was any sense that DDoS was a passing problem of 2012. The 2024 guide treats it as a standing risk for public services. [6]

Spotted an error? Tell us and we will check it and correct the article where needed.

  • DDoS
  • online banking
  • incident notification
  • FFIEC
  • operational resilience