You have an option to select pending reason levels to understand why your jobs are pending. This is done by classifying pending reasons into categories, and then showing only the top reason in the top category.
- Use Uncategorized pending reason (-p0) to get the job ID, the user, and the pending reason of each pending job in the cluster.
- Use Key pend reason (-p1) to retrieve the key pending reason for a job. If the key pending reason is a host-based reason then the key pending reason and the corresponding number of hosts is shown. Otherwise, only the key pending reason is shown. Note: If there is no host reason other than the main reason, the main reason is shown as the key pending reason for the job.
- Use Candidate host pending reason (-p2) to better understand why your jobs are not running, and what you can do about it. Option bjobs -p2 shows the total number of hosts in the cluster and the total number considered. For the hosts considered, the actual reason on each host is shown. For each pending reason, the number of hosts that give that reason is shown. The actual reason messages appear from most to least common. Host-based pending reasons are grouped into reasons of candidate hosts and reasons of non-candidate hosts, but with this option, you will see the reasons only for the candidate hosts. Reasons for candidate hosts can be used to get the job to run. For example, with the reason “Job's resource requirement for reserving resource (mem) not satisfied”, the user may choose to lessen the job's memory requirement.