[AusNOG] [AUSNog] : Re Data Centre Fire Suppression Safety

[James Troy]

I too can confirm this when a staff member ‘accidentally’ hit the gas release instead of the door release. We ended up having to replace 80% of our SAN’s disks due to both failure and predictive failure.

 

>From what I can determine the gas nozzles is what dictates the noise variable and the pressure has to be there to a) extinguish the fire and b) because the contents are under pressure and is therefore unavoidable.

 

A interesting side-effect that we noted was out Dell SAN’s had a safety cut-off that when they ‘detected’ the issue the stopped all the disks and we have zero failures in that kit. I was quite impressed by this.

 


James Troy

Senior Systems Administration



 


 


 


0412 449 074

james.troy at agilityapplications.com

 

Level 11, 356 Collins St

Melbourne VIC 3000

PO Box 2795 | New Farm QLD 4005

 <http://agilityapplications.com/> agilityapplications.com

 





 


 


 

 

 

From: AusNOG <ausnog-bounces at lists.ausnog.net> On Behalf Of Robert Hudson
Sent: Friday, 14 December 2018 7:45 AM
To: Bevan Slattery <bevan at slattery.net.au>
Cc: ausnog at lists.ausnog.net
Subject: Re: [AusNOG] [AUSNog] : Re Data Centre Fire Suppression Safety

 

I can confirm that the sound/pressure wave from a gas discharge can, does and absolutely did destroy a lot of spinning disks in some very expensive kit (big data and database appliances) when such a system was accidently deployed in a datacentre I am familiar with.

 

On Fri, 14 Dec. 2018, 5:02 am Bevan Slattery <bevan at slattery.net.au <mailto:bevan at slattery.net.au>  wrote:

It’s pretty much all been said.

 

Halon (long gone).  Reaction sucks oxygen out of air.

FM200 (safe but being phased out).  Heard it can leave a residue despite the brochure saying not.

Inergen  more common (and others like it).  Fundamentally mostly nitrogen that drops oxygen below 15% and drops temperature.  These are two components of a fire (heat, fuel and oxygen).  People can operate comfortably below 15% oxygen.  In fact at 10% you can still function more than enough to pick up your gear and leave the room.

 

I did quite a bit of research on reduced oxygen environments (hypoxic) which is used on (Firepass etc.) http://www.firepass.com/oxygen-reduction-fire 

 

Obviously dry pipe is used a lot. The issues with gas suppression today are more around noise (and vibration) and temp drop and they relate to spinning disks and circuit boards, more than people.

 

The issues around dry pipe is, well when it goes off, it’s not very dry and water/equipment certainly doesn’t mix.

 

Cheers

 

B

 

  _____  

From: AusNOG <ausnog-bounces at lists.ausnog.net <mailto:ausnog-bounces at lists.ausnog.net> > on behalf of Paul Wilkins <paulwilkins369 at gmail.com <mailto:paulwilkins369 at gmail.com> >
Sent: Wednesday, December 12, 2018 3:53 pm
To: AusNOG at lists.ausnog.net <mailto:AusNOG at lists.ausnog.net> 
Subject: [AusNOG] [AUSNog] : Re Data Centre Fire Suppression Safety 

 

Every data centre has a fire suppression system. We're not used to thinking of this as a hazardous environment, but consequent totwo techs being found dead working on a fire suppression system in Antarctica <https://www.theguardian.com/world/2018/dec/12/antarctica-two-technicians-dead-mcmurdo-station-ross-island> , I find myself wondering yet again, why there aren't more stringent controls around the fire suppression systems in data centres: viz - when you enter a data centre, how confident can you be you're not going to be quietly asphyxiated?

Kind regards

Paul Wilkins

_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net <mailto:AusNOG at lists.ausnog.net> 
http://lists.ausnog.net/mailman/listinfo/ausnog

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20181214/45cdc6fb/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 15109 bytes
Desc: not available
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20181214/45cdc6fb/attachment-0001.png>