Benjamin Allan

This should be an api extension that could be back-compatible with 4.3.

@narategithub any thoughts on this?

@nick-enoent I expect that if maestro is managing producers, then this ldms issue will affect maestro in unhappy ways.

Don't know if there's any protocol relation that would affect #771 also.

@tom95858 no, i'm seeing this with configurations that don't involve maestro in any way. Basically, if i configure an aggregator and then later send the statement list that inverts the...

@nichamon With procstat fixed, i'm still sometimes seeing this producer message. much more frequently with valgrind in action than without, which suggests it is indeed some timing issue. What i...

The scenario being tested is that the aggregator is sent the revconf.1 script to tear it down, then the 2 samplers are deconfigured by being sent (e.g. many/run/revconf.2). Are you...

Will run more diagnostics to see if this looks a lot like #771 at the L1.

The symptom is repeatable, but a bit arbitrary, thusly: - run pid sampler on 20 nodes. - launch 36 ranks/node of mpiGraph using mpiexec or srun. - job completes; tracked...

It appears this issue has been resolved by #1153.