Applicable Delphix Versions
- Click here to view the versions of the Delphix engine to which this article applies
Major Release All Sub Releases 6.0 188.8.131.52, 184.108.40.206, 220.127.116.11, 18.104.22.168, 22.214.171.124, 126.96.36.199, 188.8.131.52
Troubleshooting why jobs are going into queued status
There are some jobs already running. Executing further jobs result in these jobs going into 'queued' status, as per the example below.
In the Job Monitor Screen (Home > Monitor):
In this example, the masking engine has been allocated 64Gb RAM.
The Monitor Screen displays the total amount of memory that can be used by jobs (44 Gb) and how much of that memory was in
use by the three currently running jobs (33.7 Gb).
The two queued jobs had been defined to run with Job min memory and max memory values as below
Job id 45, with job queue position 1:
Job id 334, with job queue position 2:
With the min memory as above, essentially this means that the first queued job (Job 45) would try to reserve right at the beginning of its job execution 16Gb,
the second would reserve 1Gb .
With the current running jobs already using 33.7 Gb out of the 44 Gb total memory for job executions, the result is that:
- Job 45 was queued because only 10Gb of 'available memory for Jobs', is available.
- Job 334 was queued even though there is 10Gb of the total jobs memory available.
The current Job Queue mode is unable to reorder or prioritize queued job positions. The queue follows a 'first in, first out' model, so it waits for 1363 to get executed/started off first.
At the time that the 2 jobs (the queued jobs) were submitted, the Engine determined that 33.7 Gb of memory (of the Memory available to Jobs) was being used
from a pool of 44 Gb. This means that any jobs submitted at that point can not individually or collectively have a Min Job Memory exceeding 10248 Mb (10Gb) .
To avoid job executions becoming queued the options are:
1) Increase the available memory to the Masking engine.
2) Determine the optimal Min job memory requirements for each job.
3) If running several jobs simultaneously, check the current available total memory for jobs before running more jobs.