Wednesday, July 02, 2008
Issue with cluster 1
We are currently experiencing an issue with web hosting cluster 1. Cluster 2 and 3 are operating within tolerances if a little slow.
We are currently looking into storage issues, as yesterday an air conditioning failure in the data centre IFL2 resulted in a sizable heat spike for a prolonged period.
UPDATE 17:00
The cause of the issue has been traced to the file server and the underlying cause identified as an overheating CPU. We believe this is associated with failure of some air conditioning units in IFL2 yesterday afternoon.
We will be working on the racks over the next week to improve air flow through the servers as well as increasing our range of monitoring sensors to include more temperature sensors on servers.
We are currently looking into storage issues, as yesterday an air conditioning failure in the data centre IFL2 resulted in a sizable heat spike for a prolonged period.
UPDATE 17:00
The cause of the issue has been traced to the file server and the underlying cause identified as an overheating CPU. We believe this is associated with failure of some air conditioning units in IFL2 yesterday afternoon.
We will be working on the racks over the next week to improve air flow through the servers as well as increasing our range of monitoring sensors to include more temperature sensors on servers.
# posted by Netserve @ 3:19 PM 
