Due to a serious hardware failure[0] we've had to take much of our file storage cluster offline. This should not affect any existing toolforge or cloud-vps usage, but new file-intensive work (e.g. creating a big new database or volume) may fail due to lack of space.
We hope to have things restored to full service in a couple of days but right now we're just watching the suspect switch to see if it fails again.
Sorry for any inconvenience!
-Andrew
This maintenance is now done and cloud capacity is back to normal.
On 8/13/24 12:43 PM, Andrew Bogott wrote:
Due to a serious hardware failure[0] we've had to take much of our file storage cluster offline. This should not affect any existing toolforge or cloud-vps usage, but new file-intensive work (e.g. creating a big new database or volume) may fail due to lack of space.
We hope to have things restored to full service in a couple of days but right now we're just watching the suspect switch to see if it fails again.
Sorry for any inconvenience!
-Andrew
cloud-announce@lists.wikimedia.org