Dear Chris,
thanks for the suggestions. I'm running Centos Stream 8.4.
I've done a couple of tests:
1) I've modified as suggested this line as this
ProctrackType=proctrack/linuxproc. Restarted the slurmctld and nd the
nodes' slurmd(hope it's enough) but didn't changed the behaviour.
2) I've tried the cgroup configuration like this:
##############Lines added/changed to slurm.conf ###############
ProctrackType=proctrack/cgroup
TaskPlugin=task/cgroup,task/affinity
JobAcctGatherType=jobacct_gather/cgroup #optional for gathering metrics
PrologFlags=Contain #X11 flag is also suggested
###########Lines of cgroup.conf#################
###
# Slurm cgroup support configuration file.
###
CgroupAutomount=yes
CgroupMountpoint=/sys/fs/cgroup
ConstrainCores=yes
ConstrainDevices=yes
ConstrainKmemSpace=no #avoid known Kernel issues
ConstrainRAMSpace=yes
ConstrainSwapSpace=yes
TaskAffinity=no #use task/affinity plugin instead
Restarted the slurmctld and nd the nodes' slurmd (again hoping it's
enough) but again no luck...
Do I need a complete restart?
What ele can I check/change/try?
Hope someone can help, thanks
Andrea
Il 10/07/2021 22:54, Christopher Samuel ha scritto:
Hi Andrea,
On 7/9/21 3:50 am, Andrea Carotti wrote:
ProctrackType=proctrack/pgid
I suspect this is the cause of your problems, my bet is that it is
incorrectly identifying the users login processes as being part of the
job and thinking it needs to tidy them up in addition to any processes
left over from the job. It also seems to be more for BSD systems than
Linux.
At the very least you'd want:
ProctrackType=proctrack/linuxproc
Though I'd strongly suggest looking at cgroups for this, see:
https://slurm.schedmd.com/slurm.conf.html#OPT_ProctrackType
and:
https://slurm.schedmd.com/cgroups.html
Best of luck!
Chris
--
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Andrea Carotti
Dipartimento di Scienze Farmaceutiche
Università di Perugia
Via del Liceo, 1
06123 Perugia, Italy
phone: +39 075 585 5121
fax: +39 075 585 5161
mail: andrea.caro...@unipg.it
