Mark Grondona

Results 154 issues of Mark Grondona

The stdin shell plugin reads input data from the KVS and tries to write it to the desitination task(s), even if the tasks aren't reading stdin. If the task is...

On elcap a large job was stuck in CLEANUP with many active job shells still running. The logs indicate that a SIGKILL was sent to the shells, but this apparently...

I had inadvertently created a config with: ``` [[resource.config]] hosts = "pi[0-2]" cores = "0-3" properties = "cm4" ``` This is an error since `properties` is not an array, but...

We just hit this assertion failure on elcap: ``` broker: subprocess.c:249: subprocess_check_completed: Assertion `p->state == FLUX_SUBPROCESS_EXITED' failed. ``` a core was captured. Here's the backtrace ```gdb #0 0x00007ffff61eeacf in raise...

Problem: Just after the shell connects to the broker and a `flux_t` handle is available, the evlog based shell logger disables log output to `stderr` completely. However, at this time...

Someone mentioned that it would be useful if `flux job taskmap` had a mode to output the taskmap for a job as a set of hostname to ranks lines. This...

There are several things that currently block an orderly flux shutdown, including slow epilogs that hold jobs in CLEANUP, bugs in jobtap plugins that leave jobs needing manual cleanup, etc....

While debugging MPI abort handling I noticed that there's what appears to be an unnecessary delay in job termination when `exit-on-error` triggers: ``` bash-4.4$ flux run -vvv -N4 -o exit-on-error...

While trying to run some of the MPI tests under a test instance, I noticed that MPI bootstrap with mvapich2-2.3.7-intel hangs when there are multiple tasks per node: ``` $...

The resource module returns an errno only when the inventory isn't available. This can be confusing to admins, so perhaps a human readable error should be added for this case....