The root cause of the ticket views issue on August 7, 2024, was a process breakdown during a blue/green cluster switch. This led to an incomplete deployment, causing configuration errors in the production Pods.
To resolve the issue, the engineering team manually scaled up the indexers on the active cluster to process the accumulated ticket updates and resumed the cluster-switch procedure with the correct configuration.
On August 7, 2024, Zendesk experienced a service incident affecting ticket views. From 20:48 UTC to 23:27 UTC, customers across all Pods faced issues such as solved tickets not disappearing, slow updates, and incorrect ticket numbers displaying….
Zendesk resolved the ticket views issue by manually scaling up the indexers on the active cluster to handle the accumulated ticket updates. The engineering team then resumed the cluster-switch procedure with the correct configuration. This approach…
To prevent future ticket view issues, Zendesk plans to review the cluster switch process to ensure thorough post-deployment checks. This will help prevent incomplete deployments from going unnoticed. Additionally, Zendesk aims to develop automation…
For more information about Zendesk service incidents, you can check the system status page, which provides current system status updates. The summary of post-mortem investigations is usually posted a few days after an incident has ended. If you…