Body
Quest and Kellogg Linux Cluster Downtime, December 14 - 18.
Quest, including the Quest Analytics Nodes, the Genomics Compute Cluster (GCC), the Kellogg Linux Cluster (KLC), and Quest OnDemand, will be unavailable for scheduled maintenance starting at 8 A.M. on Saturday, December 14, and ending approximately at 5 P.M. on Wednesday, December 18. During the maintenance window, you will not be able to login to Quest, Quest Analytics Nodes, the GCC, KLC, or Quest OnDemand submit new jobs, run jobs, or access files stored on Quest in any way including Globus. For details on this maintenance, please see the Status of University IT Services page.
Quest RHEL8 Pilot Environment
The RHEL8 Pilot Environment is available for use now.
Starting Tuesday, December 10, seventy of the latest Intel Emerald Rapids CPU nodes (128 cores and 512GB of RAM) will be available to the RHEL8 Pilot, significantly increasing the compute capacity of this environment. With this expansion, the pilot environment will consist of twenty-four NVIDIA H100 GPU nodes (totaling ninety-six H100 cards) and 140 CPU nodes totaling 12,600 cores. Quest users are encouraged to test their workflows in RHEL8 Pilot environment to prepare for Quest moving completely to RHEL8 in March 2025. Detailed instructions are available on how to submit jobs for the new Operating System in the Knowledge Base article, RHEL8 Pilot Environment.
Tips for addressing common problems using R on Quest.
Common Issues
My RStudio session is frozen
If your RStudio session is frozen, one option is to delete your .rstudio and .rstudio-desktop directories in your Quest home directory.
The risks of doing this are:
- Losing any unsaved files or objects in the environment in RStudio sessions
- Disrupting any running sessions
- RStudio not remembering what project or files you had open
To delete these directories, first log out of all Quest Analytics sessions and close any RStudio sessions you have running from Quest login nodes or interactive jobs. Then, if you aren't already, connect to a Quest login node.
NOTE: Please type the following commands carefully. Removing files on Linux systems can't be undone.
cd
rm -r .rstudio
rm -r .rstudio-desktop
The try launching a new RStudio session however you were using it before. If you're still encountering a problem, contact quest-help@northwestern.edu for further assistance.
My R script runs in R but gives me errors as a batch job
Case 1: Your script runs in R or RStudio on the login nodes and as an interactive job on the compute nodes but gives you an error as a batch job
If you are using Rscript as the R executable in your batch job submission script, you may need to add library(methods) to the top of your R script (your .R file).
The R executable that runs in RStudio, or starts the R console, or is invoked via R CMD BATCH automatically loads the methods library by default. Rscript doesn't. This often leads to errors concerning object types or classes, or errors surrounding types of matrices.
Case 2: Your script runs in R or RStudio on the login nodes but you get an error on the compute nodes via both an interactive job and batch job
One of the following situations likely applies:
- A package you are using needs to access a system library that isn't installed on the compute nodes. This most commonly occurs surrounding locale information or fonts.
- You installed a package with a non-default version of a compiler module loaded. You may need to add a module load statement for the relevant compiler to your job submission script.
- You're accidentally using different versions of R in the two places. Make sure you're loading the R module with the version specified.
If you're still encountering a problem after trying the above, contact quest-help@northwestern.edu for further assistance.
I'm getting a compilation error when installing or updating a package
If you're using the Quest Analytics Nodes, you may need to log in to Quest directly to install or update the package. This is because you may need to load a newer version of system libraries via the module system to compile the package during installation successfully. See installing R packages for an example.
If you're not using the Analytics Nodes, or you're not sure what you need to change to get the package to install successfully, contact quest-help@northwestern.edu for further assistance.