OpenClear

Home / Docs / Incident notification

Incident notification

Current as of 2026-10-22. Checked against its sources at least once a year (control check regulatory_currency).

Three rules decide who hears about an incident involving OpenClear, and when. This isn't legal advice (Compliance).

Rule Who tells whom When
12 CFR 304.24 (FDIC) OpenClear's host, as a bank service provider, tells at least one contact the bank designated, or else its CEO and CIO As soon as possible after determining an incident that has disrupted or degraded the service, or is reasonably likely to, for four hours or more. Scheduled maintenance announced beforehand doesn't count
12 CFR 304.23 (FDIC) The bank tells its FDIC supervisory office As soon as possible, and within 36 hours of determining a notification incident
23 NYCRR 500.11(b)(3) (New York banks) The host tells the bank, as the contract says On a cybersecurity event directly affecting the bank's systems or data; DFS's 2025 guidance asks for "immediate or timely" notice
23 NYCRR 500.17(a) (New York banks) The bank tells DFS Within 72 hours of determining a cybersecurity incident at the bank or at its service provider
Bank Service Company Act § 7(c) The bank tells the FDIC the service relationship exists Within 30 days of the contract or the first service

Designating contacts

An administrator names the bank's incident contacts in Settings → Incident contacts, with the CEO and CIO as the fallback the rule gives, and confirms them each year. The control check Incident contacts named (control OC-IR-02) warns while there are none, while a notice is unacknowledged, or once a year has passed. Changing them is a high security event, incident_contact.changed: whoever sets them decides where notices go.

How the host notifies

From Platform → Incidents the installation's operator sends a notice to each bank affected (control OC-IR-01): - it's emailed at once to the bank's incident contacts (or its CEO and CIO, or, with neither, its administrators), whatever anyone's notification settings; - it shows to the bank's administrators on every page until one acknowledges it; - sending and acknowledging are in the bank's action log and raised as security events, incident.notice.sent (critical) and incident.notice.acknowledged, so they reach the bank's SIEM; the evidence export lists them; - updates and closure follow the first notice the same way; scheduled maintenance is announced the same way, beforehand.

A notice doesn't wait for a second operator: speed comes first, and the platform's own action log records who sent it. The runbook has the steps.

The bank's part

CUEC
B-IR-6-05 Name incident contacts, and confirm them each year
B-IR-6-06 Decide whether it's a notification incident; notify the FDIC within 36 hours
B-IR-6-07 New York banks: notify DFS within 72 hours
B-SA-9-03 Tell the FDIC about the service relationship within 30 days
B-IR-6-01 Respond to an incident that involves OpenClear

The host's part is H-IR-6-01 (notify banks), H-IR-8-01 (an incident response plan) and H-CM-4-01 (announce maintenance).

Sources

Read 22 October 2026: - 12 CFR Part 304 Subpart C, 304.21–304.24 (eCFR, current to 21 October 2026; unchanged since the rule's compliance date, 1 April 2022). - 23 NYCRR Part 500 as amended in November 2023 (DFS's own compiled text, which it marks as not the official version), and DFS's industry letter of 21 October 2025 on third-party service providers. - 12 U.S.C. 1867(c), the Bank Service Company Act § 7(c) (US Code, 2024 edition).

Unconfirmed: the FDIC's form and channel for the § 7(c) notice; whether a hosted positive pay service is a § 7(c) "covered service" (likely, as data processing; for counsel).

Built from docs/compliance/incident-notification.md in the OpenClear repository. Last changed October 22, 2026.

See positive pay from the bank's side

Start a free sandbox bank with sample checks, ACH debits and exceptions.

Start a sandbox bank