Outages and Maintenance
-
Hammer Cluster Maintenance Upgrade
The Hammer cluster will be taken down on Monday, June 4th, 2018 at 8:00am EDT for a planned upgrade to CentOS7. This process is expected to take two days, and Hammer will not return to service until 5:00pm on %enddate%. This upgrade may affect applic...
-
The Scholar cluster for educational use has proven to be extremely useful as an in-class tool. As a result of this success, it is being rebuilt and expanded over the summer. The new Scholar will offer more and better resources for educational use a...
-
The BrownGPU cluster will be unavailable beginning at Tuesday, June 26th, 2018 at 8:00am EDT, for scheduled maintenance. The cluster will return to full production by %enddatetime%. During this time, BrownGPU will have some minor configuration manage...
-
The Fortress Archive will be unavailable Wednesday, June 27, 2018 from 8:00am – 5:00pm for partial physical relocation. During this time, Fortress will receive normal updates and also have a portion of the servers which support it physically relocate...
-
Starting at Thursday, July 5th, 2018 at 8:00am EDT, Workbench will be taken down for monthly maintenance and updates. During this time, you will be unable to log in, and any active sessions or programs will be terminated. Workbench will be returned t...
-
Unscheduled Conte scratch outage
The Conte cluster began experiencing issues with the scratch filesystem around 10:45am EDT. Engineers are currently diagnosing the issue and are working to identify a fix. Job scheduling has been paused while this issue is being addressed. We will pr...
-
Old Halstead Scratch Retirement
The old Halstead and HalsteadGPU scratch storage will be retired July 9, 2018 8:00am - July 16, 2018 5:00pm EDT. This storage has not been the default scratch space on Halstead and HalsteadGPU since May 15, and its retirement will complete the pro...
-
Firebox Virtual Server Maintenance
Each of the Firebox Virtual Servers will be rebooted to have minor updates applied at some point on Wednesday, July 11, 2018 within the range of 8:00am – 12:00pm EDT This work is only for routine first-Wednesday maintenance updates, and each system w...
-
The Fortress Archive will be unavailable Wednesday, July 11, 2018 from 8:00am - 10:00am EDT for scheduled monthly maintenance. During this time, Fortress will receive normal updates and have some work done on its battery backup system. Any work which...
-
Unscheduled network outage on Brown, Rice, Snyder and Hammer
The Brown, Hammer, Rice, and Snyder clusters began experiencing issues with scratch filesystems and network connectivity around 2:30am EDT. Engineers are currently diagnosing the issue and are working to identify a fix. Job scheduling has been paused...
-
The Conte cluster will be retired at the planned end of life for this system on Wednesday, August 1, 2018. Please start archiving any data you may have there to the Fortress Archive now, or move it to other clusters or Depot for any ongoing work. Aft...
-
On Thursday, August 2nd, 2018 at 6:00am EDT, all Research Computing systems will be going offline to allow for major coordinated maintenance of central cooling systems across data centers and upgrades to the research networking core. This will affect...
-
The Rice cluster is currently experiencing issues with the scratch filesystem. Engineers have identified the problem and are in the process of applying the fix. Job scheduling has been paused while this issue is being addressed. We will provide an up...
-
Scratch unavailable on Snyder cluster
As of approximately 2:30pm EDT, the Snyder cluster is currently experiencing issues with its scratch filesystem. Engineers are currently diagnosing the issue and are working to identify a fix. Attempts to access scratch will likely fail. Job scheduli...
-
The Data Workbench and WSC Hadoop cluster began experiencing issues with connectivity around 2:00pm EDT. Engineers are currently diagnosing the issue and are working to identify a fix. We will provide an update by 3 pm.
-
Critical maintenance of RStudio Server on Scholar
RStudio Server on the Scholar Cluster will be unavailable tonight, Monday 27 Aug, 2018, from 11:00 pm to approximately midnight. During this time, a fix will be implemented for a critical problem that causes the system to not accept web-based logins....
-
The Hathi Hadoop cluster is being retired on Friday, August 31st, 2018. This was Purdue's first generally available Hadoop cluster, and it sparked further interest in other Hadoop systems. Dedicated Hadoop resources are available for research group...
-
The github.rcac server will be briefly unavailable Friday, August 31, 2018 from 8:30am - 9:00am EDT for an emergency SSL certificate update. Operations requiring connection to the central server (e.g. new check-outs or pull requests) may time out dur...
-
The Data Depot central servers lost their connection to Depot and all rebooted around 11:45am EDT. Servers are already returning, but engineers are watching the situation closely. Scheduling on clusters is being paused. Cluster jobs might be able...
-
Once again, the first Wednesday of the month, the github.rcac server will be briefly unavailable Wednesday, September 5, 2018 from 1:00pm - 5:00pm EDT for a planned maintenance. During this time, the server will undergo maintenance tasks that can not...