Rackspace datacenter infrastructure took 12-hour nap in London, Sydney, Hong Kong

Trending 3 months ago

Updated Rackspace is successful a messiness again.

The cloudy concern's position page reports outages successful its SYD2, LON5, LON3, and HKG5 datacenter infrastructure crossed May 29 and 30.

Rackspace's first incident study is timestamped 29 May 22:24 CDT.

A consequent update identified nan rumor arsenic related to Dense Wavelength-Division Multiplexing (DWDM) successful London, arsenic that installation is related to a fibre carrier web that allows Rackspace to present postulation betwixt datacenters and net work providers.

But an hr later Rackspace ruled retired DWDM arsenic a origin of nan incident. The institution has not updated its position page since.

The Register has obtained an email a SaaS institution that resides successful Rackspace has sent to its customers.

"Our hosting supplier Rackspace person confirmed they are experiencing connectivity issues," nan email opens. "All disposable engineers person been engaged and are moving to resoluteness nan rumor pinch nan highest priority."

It gets worse: Rackspace has warned customers of its London datacenters that whatever's causing nan rumor whitethorn disrupt their backups, and offered instructions connected really to observe immoderate failures.

At nan clip of penning – 02:45 CDT connected May 30 – Rackspace had not updated its position page for complete an hour. The Register has sought remark and will update this communicative if we person useful information.

  • Rackspace racks up occupation cuts amid marketplace downturn and talk of offshoring
  • Rackspace confirms ransomware onslaught down days-long email meltdown
  • On nan 12th time of nan Rackspace email disaster, it did not springiness to maine …
  • Microsoft ditches plans for 500,000 sq ft London office

This outage comes astatine a unspeakable clip for Rackspace arsenic its US and UK customers look from a vacation weekend.

The institution is besides acold from retired of nan woods aft nan December 2022 attack connected its Hosted Exchange situation caused weeks of disruption and saw nan work abandoned.

That incident led to prolonged inability to entree data, again pinch unspeakable timing arsenic customers prepared for nan festive season. Class actions are nether measurement to springiness aggrieved customers a chance for compensation.

And now Rackspace customers connected 3 continents person a caller group of worries. ®

Updated astatine 23:00 UTC, May 30

Rackspace has identified nan origin of nan problem arsenic "I/O limits successful nan multi-tenant Shared SAN situation had reset incorrectly."

Rackspace ran a book to reset nan worth and arsenic of 12:10 CDT services were restored – pinch immoderate exceptions.

"It has been identified that immoderate impacted Linux VMs (virtual machines) will not automatically retrieve if retention has been adjusted and will request to beryllium manually rebooted. Rackspace engineers tin reboot impacted VMs from nan portal wherever necessary" states Rackspace's position update.

A Rackspace spokesperson told america nan incident is not considered a information matter.