Buildfarm failure for `v0.38.0`

[jameshcorbett]

Failure in t3034-resource-pconstraints.t:

80: /builddir/build/BUILD/flux-sched-0.38.0/t/sharness.sh: line 341: 1529346 Aborted                 (core dumped) ${query} -L hetero.json -f rv1exec -S CA -P hinodex -t 008.R.out -o 008 -g graphml -F rv1_nosched < cmds008
80: not ok 10 - graphml output can be generated with properties

[grondo]

Did the coredump get collected by the gdb command in the %check section? (If that's still there)

[grondo]

Oh it isn't getting to the gdb part because this line is failing:

  (cat t/*.output t/*.log &&

+ cat 't/*.output' 't/*.log'
cat: 't/*.output': No such file or directory
cat: 't/*.log': No such file or directory
error: Bad exit status from /var/tmp/rpm-tmp.0ppOss (%check)

If you remove that line maybe we can get a backtrace?

[jameshcorbett]

Yeah I just fixed that and am re-running

[jameshcorbett]

106/106 Test  #44: t1024-alloc-check.t ..........................................   Passed   59.94 sec
99% tests passed, 1 tests failed out of 106
Total Test time (real) =  59.99 sec
The following tests FAILED:
     80 - t3034-resource-pconstraints.t (Failed)
Errors while running CTest
Output from these tests are in: /builddir/build/BUILD/flux-sched-0.38.0/Testing/Temporary/LastTest.log
Use "--rerun-failed --output-on-failure" to re-run the failed cases verbosely.
+ echo -e 'thread apply all bt\nquit'
++ find . -name '*.core'
+ exit 1

No luck it seems

[grondo]

Oh no are corefiles being captured by something now in the build farm? I'm afraid you might have to try recreating outside of the build farm.