1579118 : Your held Jobs

Created: 2026-02-24T07:36:01Z - current status: new

Here is the anonymized and summarized version of the reported issue:

Summary of the Issue

A large number of jobs ([~500]) were placed on hold due to exceeding the requested disk space allocation. The jobs had a RequestDisk value of 0 MB, but their actual disk usage (DiskUsage) reached 1500 MB. Since the jobs did not explicitly request disk space, the system enforced a default limit, causing them to be held.

Solution

  1. Option 1: Remove the held jobs Delete the jobs and resubmit them with updated disk requirements.

  2. Option 2: Edit and release the jobs

  3. Update the RequestDisk attribute to a value greater than the actual disk usage (e.g., RequestDisk = 2000 for 1500 MB usage).
  4. Use the following commands: bash condor_qedit [JOB_ID] "RequestDisk = 2000" condor_release [JOB_ID]
  5. For bulk editing, loop through all held jobs or use wildcards (if supported).

Prevention for Future Submissions

  • Explicitly define RequestDisk in the job submission file to avoid default limits.
  • Monitor disk usage during job execution to ensure requirements align with actual needs.

Sources Referenced

  1. NAF Documentation: Job Requirements
  2. NAF Cheat Sheet: Checking and Managing Submitted Jobs