Currencycloud Outage History

# **Overview** A high volume of RON payments experienced delays in status updates, remaining in a “submitted” state longer than expected. The issue primarily affected larger batch submissions. # **Client Impact** Payments were processed as expected however, status updates were delayed, causing transactions to remain in a “submitted” state longer than usual. # **Root Cause** The delay was caused by processing constraints on the side of our network partner following increased volumes after a bank holiday, which impacted the timely issuance of status updates. # **Remediation** We confirmed that payments were successfully received and processed by our network partner and continued to monitor progress to completion. We are working closely with the partner to improve the timeliness of status updates and reduce the likelihood of similar delays during peak volume periods.

# **Overview** On 2 April 2026, the production API experienced a temporary service disruption, resulting in increased request failures and timeouts. The issue was identified and resolved following the rollback of a recent internal change. # **Client Impact** Clients experienced intermittent failures, including authentication issues and timeouts when accessing the production API, leading to temporary service disruption. # **Root Cause** The incident was caused by an internal deployment change that impacted connectivity between the API Gateway and upstream services, affecting request processing. # **Remediation** The impacted change was rolled back, restoring normal service. A full root cause review is underway, with improvements being made to deployment controls and validation processes.

# **Overview** An issue impacted the processing of certain DKK payments, where payments did not progress as expected despite funds being successfully received by our DKK network partner. # **Client Impact** Some payments experienced delays in completion. The majority were processed on the same day, with a small number completing the following business day. # **Root Cause** The delay was caused by a processing gap on the side of our DKK network partner, where incoming funds were not promptly applied, preventing payments from completing as expected. # **Remediation** The issue was resolved following intervention with our DKK network partner to apply the funds and complete processing.

# Overview The inbound Confirmation of Payee \(COP\) service experienced slower‑than‑expected processing of account updates. While core COP checks continued to operate normally, some newly created accounts took longer to appear in the directory service of an external partner, which could affect matching availability for some customers. Normal processing has since been fully restored, and the backlog of updates has cleared. # Client Impact New accounts took longer than expected to appear in the external partner directory. Some sub‑accounts were also delayed in being registered. Existing accounts were not affected and continued to receive expected COP match responses. Core COP request/response behaviour remained within expected performance levels. # Root Cause The delays occurred because the COP inbound service received a higher‑than‑usual volume of account updates over a short period. These updates are sent to an external partner’s directory service, which applies standard processing limits to ensure stability. The temporary increase in traffic exceeded those limits, leading to a queue of updates waiting to be processed. As a result, some newly created accounts took longer than usual to appear in the directory. Once the update volume naturally returned to normal levels, the queue cleared and processing performance stabilised without further action required.

# **Multiple issues on API DEMO** ## **Overview** Between 11–12 March 2026, some users of the API Demo environment experienced difficulties signing in and accessing certain test endpoints. This occurred during routine improvements to the demo gateway. ## **Client Impact** Affected users saw messages such as invalid credentials or 429 rate‑limit responses. This only affected testing activity in the demo environment for a limited period. ## **Root Cause** The situation arose when the Demo environment briefly began using a newer gateway that expected a different token format. Since many users were still using the previous format, this created compatibility issues that led to the authentication and rate‑limit behaviour observed. Returning the environment to the earlier gateway immediately restored all expected functionality. Currencycloud Confidential

# Overview A certificate used by the Swift Microgateway service reached its expiry. resulting in a number of requests passing through the gateway returning with “expired certificate” error messages. # Client Impact The Swift payment tracking flow experienced temporary failures. # Root Cause The issue stemmed from the certificate lifecycle reaching its end, which only became visible once the expiration date was reached on live hosts. Once each host was updated, connectivity and authentication validated successfully.

# Overview On 2026-02-20 between 15:30 UTC and 18:34 UTC, a number of clients were unable to obtain pricing for KWD currency pairs via our FX Provider Services. All other currency pairs operated normally. The issue was promptly resolved once pricing was restored by our external provider. # Client Impact Impact was limited to KWD currency pairs. Clients requesting KWD spot or forward rates during the affected period received “No Price Available” responses. No further impact on client trades # Root Cause The incident was caused by a temporary outage at our external FX provider, which prevented them from supplying KWD pricing. Connectivity remained intact, and pricing resumed once the provider restored service. Pricing resumed at 18:34 UTC, and services are fully operational.

# **Overview** During a development rollout for the Bulk Upload feature, the new flow did not consistently retain the “on behalf of” \(OBO\) account behaviour. # **Client Impact** A very small number of records were created under the house account instead of the intended sub‑accounts during Bulk Upload activity. # **Root Cause** A recent update to the bulk‑upload process introduced a change in how account information was passed between internal services. In certain cases, this caused the system to default to the house account instead of the selected sub‑account. The issue was quickly identified and corrected. ‌ ‌ ‌ ‌ ‌ **Currencycloud Confidential**

# Overview Between 17:15 and 17:31 UTC, some FX transactions in select GBP currency pairs experienced intermittent failures affecting trade date amendments and executions. The issue was identified and resolved within 16 minutes. # Client Impact Impact was limited to specific GBP pairs, with no confirmed financial or data integrity impact. All other FX services operated normally. # Root Cause The incident was caused by a temporary disruption involving an external network partner supporting certain FX transaction flows. # Remediation Connectivity was restored following corrective measures by the partner, and services are now fully operational with continued monitoring in place.

## Overview An issue occurred following an external service configuration update, which caused disruption to the processing of some TRY and CHF payments. The incident was identified when a subset of payments failed to progress as expected due to connectivity validation errors. ## Client Impact Clients submitting TRY and CHF payments experienced processing delays during the incident window. Payments remained pending until service connectivity was restored. All affected payments have since been successfully processed to value date, and no payments were lost or duplicated. ## Root Cause The issue originated from a scheduled security certificate rotation performed by an external payment provider. The corresponding certificate updates were not applied within the payment service configuration in time, resulting in validation failures that prevented successful transmission of certain TRY and CHF payments. ## Remediation The issue was resolved by updating the relevant security certificates within the payment service configuration, restoring normal connectivity and processing. Additional measures have been implemented to strengthen certificate lifecycle tracking, improve communication routing for third-party notifications, and enhance monitoring to detect similar validation failures earlier.