November 13, 2009, 4:20AM-10:43AM. Sporadic Pegasus email login failures.

October 12, 2009, 10:51AM-2:19PM. Blackboard server interruptions due to load on the authentication servers.

October 7, 2009, early afternoon. Wireless became unavailable due to a partial controller crash which took time to identify and repair.

October 7, 2009, 8:00AM. Not a failure: pegasus' load still being excessive, we added yet another server to the cluster. This, combined with refactoring the connection between WebMail and pegasus, finally allowed pegasus to scale up to meet demand. We continue to work to make it even more responsive, and we thank you for your patience.

October 5, 2009, 10:00AM - October 6, 2009, 4:00PM. Pegasus server interruptions due to the highest load we've yet seen from WebMail and users of pegasus email. We added another server to the pegasus cluster today, which helped with the load.

October 5, 2009, 10:04AM-10:32AM. 2:18PM-3:40PM. Blackboard server interruptions because of LDAP services being slow to respond, due to very high pegasus usage this week.

October 2, 2009, 11:30AM-3:30PM. Pegasus, Wireless, and Blackboard server interruptions due to an outside company's mishap near our equipment. A critical switch was fried, and we essentially had to rewire half of our network to workaround the problem. In the middle of it all, pegasus crashed, requiring manual intervention in order to boot again; however, another machine stayed running during this time, and continued to service connections.

September 10, 2009, 12:30PM-5:40PM. The DHCP server at 165.230.81.226 stopped responding, apparently causing issues in the labs several hours later when the leases expired. It is not yet known why the lab machines didn't find and use the DHCP service at 165.230.79.226; we are investigating the issue.

August 25, 2009, 7:30AM-4:30PM. Various issues affecting both DNS servers (165.230.79.226 and 165.230.81.226) at times throughout the day caused what appeared to be a slowdown in using the network.

July 29, 2009, Midday. The DNS service at 165.230.81.226 became unresponsive when trying to perform a critical security update to the service's software. This manifested itself to most users as a slowdown in using the network.

July 29, 2009, 8:30AM-9:30AM. The DNS service on andromeda became wedged when trying to perform a critical security update to the service's software. It was worked around relatively quickly but it affected all services, including a slowdown of email delivery. Blackboard showed the "unavailable" page during this time, but to most users this manifested itself as a slowdown in using the network.

June 20, 2009, 8:55PM-9:00PM. We experienced another apparent power problem in the same rack as on June 6, which affected mail delivery until Sunday at about 2:30PM.

June 14, 2009, 1:56PM-9:07PM. The storage device for Blackboard had an equipment failure, resulting in all of Blackboard's storage to be offline. A system administrator drove in and resolved the issue.

June 14, 2009, 1:56PM-4:35PM. The storage device for Pegasus had an equipment failure, resulting in some of the pegasus servers to not have access to user data. This affected sending and receiving email, as well as file and home directory access. A system administrator resolved the problem remotely.

June 6, 2009, 8:55PM-9:00PM. We experienced a power problem in one of the racks, but it was seen by staff and the systems were recovered immediately.

March 1, 2009, 3:22AM - March 2, 2009, 5:57PM. One of the Pegasus mailservers refused to allow users to send email through it. WebMail was not affected by this.

February 17, 2009, Midday. Sporadic Web Mail Outages.

February 16, 2009, 4:30AM - 9:15AM. Blackboard Outage.

August 10, 2008, 9:33PM-11:54PM. The same rackmounted UPS tripped again, causing andromeda and the campus webserver to be powered off. The next day (Sunday) staff went in and rerouted power to prevent this from happening again.

July 28, 2008, 10:40AM. A rackmounted UPS's circuit breaker tripped, taking out a rack full of devices including andromeda and the campus webserver. It was reset quickly but andromeda is a large machine that takes many minutes to boot. Services were restored by 11AM.

January 15, 2008, 6:00AM. Not a failure: several services restarted to resolve potential security vulnerabilities.

January 11, 2008, 12:40PM. Not a failure: web services were restarted to resolve potential security vulnerabilities.