Performance Issues on xeon01

There will be no further updates here.

All future information and status updates will now be published exclusively at:

https://status.dataflair.net/incidents/1#update-2

We have identified the root cause of the high I/O delays on xeon01. The issue is related to the currently installed datacenter SSDs, which are not sufficiently suited for the actual workload requirements of this cluster. The system is currently using Samsung SSDs.

Unfortunately, this limitation could not be fully validated in advance, as the level of performance growth and resource utilization exceeded our original projections.

To restore stability and improve performance, the existing drives will now be replaced with new NVMe SSDs. This upgrade is expected to significantly improve both I/O performance and overall system stability.

Further details regarding the maintenance process will follow shortly.

We have detected a very high I/O delay on xeon01. At the moment, we suspect that the issue may be related to unusually high resource consumption caused by individual customer VMs.

We are currently analyzing the situation in detail to identify the exact root cause and take appropriate action.

Further updates will follow as soon as more information becomes available.