Environmental Monitoring and Emergency Shutdown

The system athena9 is responsible for monitoring the current environmental state and for initiating a (controlled) emergency shutdown when conditions exceed preset limits.

[edit] Emergency Shutdown Triggers

At the current time, the following events trigger a controlled shutdown:

• UPS goes on battery not for self-test and for > 60 seconds
• Any of the six rack-front temp sensors exceeds 35°C (95°F)
• Any of the six rack-rear temp sensors exceeds 65°C (150°F) [Drop this? --WRS]

[edit] Emergency Shutdown Sequence

A controlled emergency shutdown via athena9 consists of the following steps:

• Turn off compute node PDU's
• Issue 'shutdown' commands to all polyserv nodes
• Wait 120 seconds for polyserv shutdowns
• Issue 'shutdown' command to cluster head node
• Wait 60 seconds for head node shutdown to initiate
• Issue 'graceful shutdown' (??? second delay) to UPS
• Shut down self

[edit] Environmental Monitoring Webpages

http://athena9.npl.washington.edu/Graphs/