Google Cloud VPN incident affecting services

What to do

• Check whether you use Aiven in africa-south1, eu, europe-central2, europe-north1. Check your stack

  Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.

• Confirm your retries, queues and scheduled jobs caught up after the 12h 47m it lasted.

  Anything that failed between 2026-10-20 06:43 UTC and 2026-10-20 19:30 UTC may need a replay.

• Subscribe to Aiven's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=aiven works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from Aiven 8

1. investigatingOct 20, 2026, 6:43 AM UTC

   We are investigating an ongoing Google Cloud VPN incident affecting the europe-west2 and africa-south1 regions. This is impacting network connectivity between nodes and may affect the availability and performance of clustered Aiven services in the affected regions. Our team is investigating the impact and monitoring affected services. We will provide further updates as more information becomes available.

   provider

2. investigatingOct 20, 2026, 6:51 AM UTC

   We are investigating an ongoing Google Cloud networking incident. Google Cloud engineers have confirmed an issue with Virtual Private Cloud (VPC) networking affecting communication between virtual machines, including communication over internal IP addresses, IPv6, and other networking paths. ``` The incident is currently reported to affect the following regions: Africa: africa-south1 Europe: europe-central2, europe-north1, europe-north2, europe-southwest1, europe-west1, europe-west2, europe-west3, europe-west4, europe-west6, europe-west8, europe-west9, europe-west10, and europe-west12 ``` This issue is impacting network connectivity between nodes and may affect the availability and performance of clustered Aiven services in the affected regions. Our team is investigating the impact and monitoring affected services. We will provide further updates as more information becomes available.

   provider

3. investigatingOct 20, 2026, 8:44 AM UTC

   The latest information identifies europe-west2 and the europe-west2-b zone as affected locations. The incident involves Virtual Private Cloud (VPC) networking issues that can disrupt communication between virtual machines. This may affect connectivity between nodes and the availability of clustered Aiven services in the affected location. Our team continues to investigate the impact on affected services and monitor their health. We will share further updates as more information becomes available.

   provider

4. investigatingOct 20, 2026, 11:10 AM UTC

   The latest information identifies `europe-west2` and the `europe-west2-b` zone as affected locations. The incident involves Virtual Private Cloud (VPC) networking issues that are disrupting communication between virtual machines in the affected location, which is impacting the availability of clustered Aiven services — currently Aiven for Apache Kafka and Aiven for OpenSearch. We continue to monitor our services to identify any remaining impacted services and we are waiting for connectivity to be restored from Google side, and for the affected nodes to return. We will share further updates as more information becomes available.

   provider

5. investigatingOct 20, 2026, 2:10 PM UTC

   We are continuing to respond to the Google Cloud incident affecting the europe-west2 region. The underlying Google Cloud issue — a power outage in the europe-west2-b zone has been mitigated by Google, and recovery of most workloads has been in progress since 15:10 EEST. Services running in the affected region may be unavailable or degraded while recovery is in progress. We are seeing the recovery of services and are prioritizing the recovery of existing workloads. Customers may not be able to yet create new workloads. We are actively recovering the affected services, including recycling affected Kafka nodes and working through the OpenSearch and PostgreSQL services that need additional steps. We are monitoring the situation closely and will continue until all affected services are healthy.

   provider

6. monitoringOct 20, 2026, 6:10 PM UTC

   All PG and Kafka services known to be affected by the Google Cloud incident in europe-west2 have been recovered. We are currently recovering and fixing known Opensearch services. We are continuing to monitor all services in the region to confirm they remain healthy. If you are still experiencing issues with a service in europe-west2, please contact Aiven Support.

   provider

7. monitoringOct 20, 2026, 6:55 PM UTC

   All Aiven services known to be affected by the Google Cloud incident in europe-west2 have been recovered. After Google mitigated the power outage in the europe-west2-b zone, we restored the affected Aiven for Apache Kafka, Aiven for OpenSearch and Aiven for PostgreSQL services by recovering nodes or replacing them. We are continuing to monitor all services in the region to confirm they remain healthy. If you are still experiencing issues with a service in europe-west2, please contact Aiven Support.

   provider

8. resolvedOct 20, 2026, 7:30 PM UTC

   This incident has been resolved. A power outage in Google Cloud's europe-west2-b zone disrupted services running in the europe-west2 region. Nodes in the affected zone were rebooted when the zone recovered. Some affected Aiven for Apache Kafka, Aiven for OpenSearch, Aiven for PostgreSQL and Aiven for Valkey services needed extra recovery steps or node replacement before they returned to normal operation. All affected services have now been restored. We apologise for the disruption.

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.