Software Update: HTCondor 9.10.1 / 9.0.15

The HTCondor Team at the University of Wisconsin-Madison has released new feature and long-term support versions of its HTCondor workload management system. The version numbers have ended up at 9.10.1 and 9.0.15. HTCondor focuses on managing compute-intensive tasks and can distribute them over several connected nodes. The user sends their task to HTCondor, which then handles the process based on set policies and the availability of connected resources, and finally sends the results back to the user. For example, HTCondor can control a dedicated Beowulf cluster, but also regular desktops that have nothing to do for a while. During the day SC16 Google, Fermilab and the HTCondor Team have a 160k-core cloud-based elastic compute cluster demonstratedand in 2020 the National Science Foundation selected HTCondor as part of her Partnership to Advance Throughput Computing. The brief adjustments to these editions are as follows:

Version 9.10.1 – Feature Channel

• ActivationSetupDuration is now correct for jobs that checkpoint

Version 9.10.0 – Feature Channel

• With collector administrator access, can manage all HTCondor pool daemons
• SciTokens can now be used for authentication with ARC CE servers
• Preliminary support for ARM and POWER RC on AlmaLinux 8
• Prevent negative values ​​when using huge files with a file transfer plugin

Version 9.0.15 – Long Term Support Channel

• Report resources provisioned by the Slurm batch scheduler when available

Version 9.0.14 – Long Term Support Channel

• SciToken mapping failures are now recorded in the daemon logs
• Fix bug that stopped file transfers when output and error are the same
• Ensure that the Python bindings version matches the installed HTCondor
• $(OPSYSANDVER) now expand properly in job transforms
• Fix bug where context managed Python htcondor.SecMan sessions would crash
• Fix bug where remote CPU times would rarely be set to zero