Users of Google Compute Engine's Persistent Disks in the europe-west1-b region have endured an anxious few hours, as the service has experienced a lengthy brown-out.
Persistent Disks are data stores that exist independently of virtual machines and retain data whatever the state of the VM. One can store data in a Persistent Disk and connect that Disk to a virtual machine of one's choice. Google offers the disks in spinning rust and flashy variations.
Google first reported the issue at 11:14 on Thursday, US Pacific time, which translates to 18:15PM GMT.
11 hours later, Google's status page suggests the service is still struggling.
Google says the problem meant “Customers who have machines running in this zone [europe-west1-b] may see read errors” and by 13:00 Pacific Time said only one per cent of machines were having problems as “the service has been partially recovered.”
Restoring from snapshots was advanced as a workaround for those experiencing the problem.
At the time of writing, Google's most recent update was posted at 22:56 Thursday Pacific Time and offered nothing more than news that “We are still working on restoring the service of Google Compute Engine Persistent Disks in europe-west1-b.”
It's been a tricky week for Google, which last weekend reported “High error rate of requests to Google Cloud Storage” for over two hours. During the week, the company's Search API for Google App Engine experienced 40-minute wobble attributed to “increasing latency started causing Search API timeouts.” ®