Outages and Maintenance
-
LustreD (/scratch/conte) unavailable
UPDATE: LustreD has been returned to service and scheduling has been resumed as of about 4 pm Saturday, October 4th, 2014. The LustreD filesystem, serving the Conte cluster, has become unavailable as of about 10:55 am Saturday, October 4th, 2014. Sys...
-
Fortress HPSS Software Upgrade
Update The issues with the HPSS upgrade have been resolved with assistance from the vendor and the system is now in normal operations. Anyone using HSI or HTAR from their personal systems will need to upgrade their client to the latest (version 5) c...
-
Network Maintenance for Carter, Conte, and Hansen clusters
The Carter and Conte clusters will be briefly unavailable on Monday, October 13, 2014 for upgrades to the clusters' respective network routers. This upgrade will significantly upgrade Conte and Carter's network connectivity, as part of ITaP's 2014 re...
-
Update: the vendor has identified and confirmed a bug in the system's software. The engineers continue assessing the situation and are working on a plan of actions. The outage window is further extended. Update: The Depot maintenance ran into some pr...
-
Diminished network connectivity to research computing resources
Due to a network issue at the Indiana GigaPOP, connectivity to RCAC resources from off campus is intermittent. Access to the research computing web site, Globus, Thinlinc, or other research computing resources may be impacted. To work around this iss...
-
Power has been restored and the Peregrine 1 cluster has been restarted and is in full production mode The Calumet campus is experiencing a power issue that has rendered Peregrine1 unavailable. Contractors are on site working to restore service, but t...
-
The Rossmann cluster will be unavailable on Tuesday, December 2, 2014 for maintenance on the cluster's network interconnect, and for routine operating system patches. Any PBS jobs which request a walltime which would take them past 6:00am on Tuesday,...
-
ECN services outage - ITaP Research Computing systems impacted
Engineering Computing Network (ECN) in coordination with Physical Facilities will be conducting a planned power outage in the MSEE building from 8am until noon on Saturday, January 10th, 2015. No one will be allowed to enter the building during this...
-
Unscheduled Outage to Conte Scratch
Update 2 Storage engineers have brought Conte scratch back to full production. Job scheduling on Conte has been restored as of approximately 2:20 pm on 1/21/2015. Any job that was actively writing to scratch while the filesystem was in a degraded sta...
-
Important Operating System Updates - Hathi Hadoop Cluster
On the morning of Tuesday, February 3, 2015, Hathi login nodes will be rebooted to apply an important Red Hat Linux operating system update.
-
Job scheduling problems on Radon Cluster
UPDATE 4:00 pm Tuesday 3 February 2015 The problem referenced below has been diagnosed and corrected, and job scheduling should work as expected on the Radon cluster. We do not believe that running jobs were affected during this outage, only the abi...
-
Important operating system updates - Community Clusters
On the morning of Thursday, February 5, 2015, Carter, Conte, Hansen, Peregrine1, Radon, and Rossmann login servers will be rebooted to apply an important Red Hat Linux operating system update. Additionally, during this time scratch storage servers w...
-
Security updates for Conte and Radon scratch
Conte scratch has returned to full production as of approximately 2:15 pm. Update: February 26, 2015 1:20pm Radon scratch has returned to full production as of approximately 10:30 am. The work on Conte scratch is progressing as planned. Original mes...
-
Research Data Depot Security Updates
As of 3:15 pm the maintenance is complete and Research Data Depot is returned to full production. Original message: The storage servers powering the Research Data Depot will undergo maintenance on Thursday, February 26, 2015 from 10:00am - 4:00pm EST...
-
ROSSMANN IS ALIVE Rossmann Cluster is now back online. Kudos to the maintenance team and thanks for your patience! UPDATE 17 March, 2015 Our systems engineers are hard at work to bring Rossmann Cluster back to life. We shall e-mail everyone as soon a...
-
The Radon cluster will be unavailable on Tuesday, 17 March, 2015. On that day, Radon's job scheduler will be updated, and its network connection to its scratch system reconfigured. As of 5:30pm Tuesday, March 17, Radon cluster has been returned to se...
-
Hathi Hadoop cluster planned outage
The Hathi Hadoop cluster will be unavailable Monday, 13 April, 2015 from 9:00 am to 1:00 pm. During that time, the cluster hardware will be upgraded with new network interfaces. The cluster will go offline at 9:00 am, and we expect the work to be com...
-
UPDATE As of 9:00 pm Tuesday, 14 April, 2015, the Conte cluster is back in full production mode. During the maintenance, all nodes were checked for reliability and system software installations were checked for consistency between nodes, and issues...
-
UPDATE As of 4:45 pm Tuesday, May 19, all the work noted below has been completed and both Hansen and Peregrine-1 have been returned to full service. Thanks for your patience. =-=-= The Hansen and Peregrine-1 clusters will be unavailable beginning at...
-
Fortress Samba service has been restored as of 10:15am on Monday, June 8th. We apologize for any inconvenience this has caused and thanks for your patience. Beginning Friday afternoon, the Fortress Samba mounts became unavailable due to an issue with...