New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
OVHcloud fire report: SBG2 data center had no extinguisher, no power cut-out
MeAtExampleDotCom
Member
in General
https://www.datacenterdynamics.com/en/news/ovhcloud-fire-report-sbg2-data-center-had-wooden-ceilings-no-extinguisher-and-no-power-cut-out/ - apparently it took three hours to shut off power due to redundant supplies and them not having an easy cut-off. A case illustrating where resilient systems designers need to consider what would happen when they actually want something to turn off… Given that, and the resulting size of the blaze, I'm not sure a lack in the fire extinguisher department would have made much difference.
OHV not commenting at this stage, their own report still being in the works.
Further discussion that might contain further relevant info at https://news.ycombinator.com/item?id=30763945 (which is where I noticed the report).
Comments
Interesting, I thought fire extinguishers are standards for data centers. Apparently not.
did they have appropriate insurance ?
The datacenter status is not available here at the moment vms.status-ovhcloud.com/sbg/index.html
The datacenter uptime links may be found below:
http://vms.status-ovhcloud.com/index_sbg1.html - SBG1
http://vms.status-ovhcloud.com/index_sbg2.html - SBG2
You can see that all servers are offline at the moment.
Sacred
12/8/2021 9:41:01 AM
I think this is deprecated :P
This really begs the question, are the other OVH datacenters in the same situation? SBG3 is supposed to be identical to SBG2 in construction and buildings like RBX1-5 are older than SBG2.
In theory an insurance company would have surveyed the site and at least make them have extinguishers in place, so maybe not.
Always have a backup.
Download your personal disaster recovery plan.
And, monitor temperature, if it goes over certain levels, initiate failover, fire may be imminent.
In a different location. Preferably with a different provider. Have as few single-points-of-failure as possible.
Yes they had so after fire in DC their shares went up in price, so conclusion is fire was very profitable more than their fire sale 😂😂
I see ... did they offer any compensation for customers
such disasters are rare though, I remember leaseweb had a fire in their NL DC in 2013, but they were very good at recovering in less than 2 days.
I think they did 1-2 months of free clean server without any data in different DC.