Yandex Cloud incident

Load balancers unavailability in ru-central1-a zone

Major Resolved View vendor source →

Yandex Cloud experienced a major incident on April 30, 2020 affecting Network Load Balancer and ru-central1-a, lasting 18m. The incident has been resolved; the full update timeline is below.

Started
Apr 30, 2020, 08:20 AM UTC
Resolved
Apr 30, 2020, 08:38 AM UTC
Duration
18m
Detected by Pingoru
Apr 30, 2020, 08:20 AM UTC

Affected components

Network Load Balancerru-central1-a

Update timeline

  1. investigating Apr 30, 2020, 08:35 AM UTC

    We noticed some problems with load balancers in the ru-central1-a availability zone. The developer team is notified and currently investigating this issue.

  2. identified Apr 30, 2020, 08:50 AM UTC

    We have identified the root cause of this issue and fixed it, all services are operating normally.

  3. resolved Apr 30, 2020, 08:56 AM UTC

    Issue was solved at 11:38 (GMT+3). On April 30, 2020, from 11:20 to 11:38 (GMT +3) due to the release of the load balancer control plane, load balancers were unavailable, and users might witness false-negative UNHEALTHY statuses of the target resources. Unavailability was caused by the auto-recovery mechanism, which removed announcements for health checkers at the time of obtaining the full state of the balancing rules in the data plane. At 11:38 (GMT +3), we entirely fixed the problem. On the side of the load balancer, verification has been improved, which led to the early withdrawal of announcements.