William Hobbs
William Hobbs
On Slack @grondo seemed to think this might be a bug in flux-job attach's status line: ``` (s=2,d=1) tioga39 ~/ext4-blocks-in-lustre/flux (flux-work)$ flux run -n1 -c1 --urgency=0 sleep inf 18.197s: job.exception...
This came up today after the fluxion meeting as I was describing my work with ephemeral and persistent ext4 files in lustre that are mounted/unmounted as a user's job starts/finishes....
Sort of a head scratcher, for a few days now, t2406 job-exec: kill-timeout > original value has been failing: ``` expecting success: flux module stats job-exec | jq .jobs.${jobid}.kill_timeout &&...
Nick Sly gave a presentation to LC this morning about building containers for TOSS 4 and TOSS 5. In this presentation, he mentioned that his development on these containers for...
Problem: the current fluxion policy matching factory doesn't allow for customizable policies. Add a custom option and refactor existing policies to behave similarly. Add unit testing for the string parsing...
This is sort of a meta-issue: a user mentioned to me that it would be nice if they had more examples of how to customize output from Flux commands. An...
```python3 import flux, flux.job f = flux.Flux() compute_jobreq = JobspecV1.from_command( command=["sleep", "0"], num_tasks=i, num_nodes=1, cores_per_task=1 ) id = flux.job.submit(f, compute_jobreq) # When we submit, we get back the job identifier...
At dinner with @trws last night, we discussed the need as Flux grows for a few additional tutorials that go beyond the traditional user training we've had in the past:...
Resident PMI expert @garlick mentioned last week: > At one point I considered adding support for simple PMI over a unix domain socket since mpich supports that, or did at...
A couple feature requests for batch jobs have come up recently: 1. An option to specify a maximum amount of time a job could be queued, and if the time...