DataFlair incident

Degraded performance on epyc460

Minor Resolved View vendor source →

DataFlair experienced a minor incident on December 26, 2024, lasting 1d 12h. The incident has been resolved; the full update timeline is below.

Started
Dec 26, 2024, 01:58 AM UTC
Resolved
Dec 27, 2024, 02:07 PM UTC
Duration
1d 12h
Detected by Pingoru
Dec 26, 2024, 01:58 AM UTC

Update timeline

  1. identified Dec 26, 2024, 01:58 AM UTC

    Duplication of disks, which has led to a major problem in the management of KVM servers

  2. monitoring Dec 26, 2024, 02:01 AM UTC

    After informing our customers in an email three days ago, we had to bring forward the maintenance work on epyc460 at short notice. All customers will be migrated to an emergency system so that we can complete the maintenance work without causing too much downtime.

  3. monitoring Dec 26, 2024, 11:33 PM UTC

    The maintenance work on the host system has been completed. All VMs have now been migrated back.

  4. resolved Dec 27, 2024, 02:07 PM UTC

    The maintenance work was successfully completed. All servers that were migrated to the xeon516 as a fallback host have been migrated back to epyc460.