Building 240 Data Center Outage, 5PM Sep 6 – Sep 9.

The building 240 Data Center cooling outage is set for Saturday, September 7 through Monday, September 9, inclusive. The expectation is that the work will be complete on Sunday, September 8, but we have to allow a day as a contingency. The building’s water-cooling system was taken down just prior to the construction of the building extension, and since then we’ve been running on the lab’s chilled water supply. This was a temporary arrangement, and we now need to revert to due to capacity issues related to other systems coming online. This work requires that the cooling for the data center will be offline during this time while the building management brings the relocated cooling system back online.

This affects the entire MCS Linux environment, including desktop workstations and home directories. It also affects the BIO file services (Y drive and X drive for linux computers). We will be powering these systems down beginning at 5PM on Friday, September 6 in order to be fully shut down for the early morning shut-off on Saturday. We advise you to move what data you need into Box before the work takes place so you can access it during the outage if required.

Services we provide which do not use your MCS home directory will not be generally affected. Some web services may see a brief service outage over the weekend as we take the opportunity to migrate to newer hardware, though we will keep these as low-impact as possible.
It is the hope of CELS Systems that if the work is completed in the expected window, we can start bringing computers and servers back online Monday morning in short order. However, it may not happen until later in the day depending on the state of the cooling system work. For planning purposes, the safe bet is to presume the workstation environment may not be available until Tuesday, September 10, though we will do everything in our power to make sure that’s not the case. We will send an announcement if we will not make the Monday morning window.

Expect other announcements from the operators of affected systems such as LCRC (Bebop/Blues), ALCF, Magellan, and others. We will send a reminder to this list prior to the outage. Thank you.