Posypała się infrastruktura w firmie OVH, która dostarcza nam hosting. Takie są ich oficjalne tłumaczenia:
Dear customer,
We host the three million websites on our web hosting packages in two datacentes: the DC in Paris (P19) and the new DC in Gravelines (GRA1). In P19, we use different technologies to host website and database data. In most cases, we use our NAS or NAS-HA technology based on ZFS.
In addition, in P19, in some cases we use EMC VNX 5400 storage bays with SSD disks. This is a solution which we put in place in 2012 to deal with storage performance issues which we experienced on some databases in 2012. Since we fixed the performance issues on our NAS-HA and in GRA we have only been using our internal solutions.
On the 29th of June at 16:30 UTC , there was an incident with one of the EMC VNX 5400 storage racks that we use to store some web hosting databases in P19. 96 SSDs configured as active/active on several physical racks were impacted. They could not be restarted. We have contacted the manufacturer and we are trying to find a solution to recover the data hosted in this rack.
This storage rack is used for hosting web hosting databases. All databases are backed up every 24 hours on other storage systems, located in another DC in Roubaix: RBX1.
We sincerely apologize. The last incident of this magnitude was in 2006 and at that time we reviewed all the storage technologies that we used at OVH. æThis outage is another lesson and we'll communicate the changes that will be made following so that we don’t see another incident of this magnitude in 10 years’ time.
Zwróćcie uwagę, że nie była to żadna awaria, a tylko "incydent" ;-)
Mam nadzieję, że forum będzie działać bez przerw technicznych przez następne 10 lat!
We sincerely apologize!
