Ticket Number: HEA-NOC/20060506-2 Ticket Status: UPDATE
Ticket Type: unscheduled Resolver: HEA-NOC
Ticket Opened: 20060506 12:03 UTC+1 Problem Start: 20060505 20:00 UTC+1
Ticket Update: 20060512 19:13 UTC+1 Problem End:
An airconditioning failure in HEAnet's client co-location room has resulted
in an unsafe operating temperature for equipment. Any affected equipment
has automatically shut down. HEAnet main comms room is unaffected.
Loss of service to all co-located equipment, including e-grants, epa,
20060506 11:54 AH Attend onsite, confirm diagnosis, work on reducing
temperature and contact air conditioning support. Equipment will not be
powered on and checked until a safe operating temperature has been
20060506 12:43 AH Additional emergency cooling and airflow has been
deployed. Temperature has dropped to safe operating temperature for some
equipment but remains excessive. Surplus equipment is being powered off.
Air conditioning support indicate possible difficulties with a callout
today. Under that circumstance, equipment will need to be powered down for
20060506 14:59 NM Air Con service engineer attended site. Expelled air of
AC units at normal operating range. NFF. Temp set for 16C.
Related Issue - Temp alarm for ircset-cpe router stopped.
20060506 15:29 NM Updated Kerna on ircset-cpe alarm and contacted Colm
Buckley ref the linux.ie box. Advised to reboot unit (power on/off).
20060512 09:21 AH All affected hosts restored. No further interruption to
20060512 16:59 BB This problem has reoccurred. Escalation to Cross
Refrigertion has been initiated. Power cycling the unit temporarily works
around the problem.
20060512 19:03 JH The air conditioning controller card was not powering
one of the two fans. This caused the pressure from the one running fan to
run very high and triggered the errors and shutdowns. The Cross
Refrigeration engineer has bypassed the controller card to power the
second fan directly. This has lowered the pressure and the air
conditioning is now functioning normally. The temperature in the room has
returned to normal. I will continue to monitor.
This ticket can be monitored at http://www.hea.net/tickets/20060506-2