Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
HTConfig
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Container registry
Model registry
Operate
Environments
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
EKPCondor
HTConfig
Commits
5523bd9c
Commit
5523bd9c
authored
10 months ago
by
Matthias Schnepf
Committed by
Matthias Schnepf
10 months ago
Browse files
Options
Downloads
Patches
Plain Diff
Add singularity test site
parent
d926c25b
Branches
Branches containing commit
No related tags found
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
0220.singularity_test.cfg
+88
-0
88 additions, 0 deletions
0220.singularity_test.cfg
scripts/config_sites.json
+26
-1
26 additions, 1 deletion
scripts/config_sites.json
with
114 additions
and
1 deletion
0220.singularity_test.cfg
0 → 100644
+
88
−
0
View file @
5523bd9c
# bwForCluster OpenStack Worker Setup
# -----------------------------------
# Cloud attributes
STARTD.CloudSite
=
"singularity"
# Matchmaking advertisement
STARTD.ProvidesIO
=
TRUE
STARTD.ProvidesCpu
=
TRUE
# localhost's condor is admin
ALLOW_ADMINISTRATOR
=
$(ALLOW_ADMINISTRATOR),
\
condor@$(UID_DOMAIN)/$(IP_ADDRESS)
# Firewall may kill connections, which are inactive for too long
TCP_KEEPALIVE_INTERVAL
=
4 * $(MINUTE)
# Wall-time
WalltimeDefault
=
54000
# Read/set MachineMaxWalltime & MachineStarttime (set by cloud-init)
include
command
into
$(ETC)/walltime.cfg
:
python
$(ETC)/meta_walltime.py
MachineWalltime
=
$(MachineMaxWalltime:$(WalltimeDefault))
MaxShutdownTime
=
$(MachineStarttime) + $(MachineWalltime)
STARTD.MaxJobRetirementTime
=
$(MaxShutdownTime) - $(JobStart:0) - 3 * $(MINUTE)
# Auto Shutdown
# Shutting down MASTER will trigger shutdown to all other daemons
# Early "nice" shutdown 10 Minutes before limit is reached
MASTER.DAEMON_SHUTDOWN
=
( TimeToLive <= 10 * $(MINUTE) )
# Fast shutdown, 3 minutes before wall-time limit is reached
MASTER.DAEMON_SHUTDOWN_FAST
=
( TimeToLive <= 3 * $(MINUTE) )
# Machine/Job environment
# Use TardisDroneUuid to identify drones in OBS
TardisDroneUuid
=
"$ENV(TardisDroneUuid)"
STARTD_NAME
=
$ENV(TardisDroneUuid)
SINGULARITY_JOB
=
TRUE
# Forces all jobs to use the CernVM-based image.
SINGULARITY_IMAGE_EXPR
=
"/cvmfs/unpacked.cern.ch/registry.hub.docker.com/mschnepf/slc7-condocker:latest"
# Maps $_CONDOR_SCRATCH_DIR on the host to /srv inside the image.
SINGULARITY_TARGET_DIR
=
/srv
# Set Scratch to default
MOUNT_UNDER_SCRATCH
=
# /scratch/.dist/etc/cvmfs:/etc/cvmfs fakes CVMFS configuration for CMS SAM tests
SINGULARITY_BIND_EXPR
=
"/cvmfs:/cvmfs /scratch/.cvmfsexec/dist/etc/cvmfs:/etc/cvmfs $ENV(HOME)/software/T1_DE_KIT/KIT-HOREKA:/etc/cvmfs/SITECONF /etc/hosts /etc/localtime"
SINGULARITY_EXTRA_ARGUMENTS
=
"--userns --env SINGULARITY_BIND= --env APPTAINER_BIND= --env APPTAINER_BINDPATH= --env SITECONFIG_PATH=/cvmfs/cms.cern.ch/SITECONF/T1_DE_KIT/KIT-T3"
# Inverse wall-time job ranking
# - prefer long running jobs (shorter jobs can back-fill)
###
# -> Scale by ( wall-time / remaining machine lifetime )
# Cut off by (START = FALSE) if RequestWalltime > TimeToLive.
# Thus we can safely set it to maximum in this case to prevent preemption
###
RankJobWtime
=
\
ifThenElse( isUndefined(TARGET.RequestWalltime),
\
0,
\
ifThenElse( TARGET.RequestWalltime > $(TimeToLive),
\
$(RankJobWtimeMax),
\
$(RankJobWtimeMax) * ( TARGET.RequestWalltime / $(TimeToLive) )
\
)
\
)
# Average CPU utilization
STARTD_PARTITIONABLE_SLOT_ATTRS
=
$(STARTD_PARTITIONABLE_SLOT_ATTRS), CPUsUsage
AverageCPUsUsage
=
Sum(My.ChildCPUsUsage)/Sum(My.ChildCPUs)
STARTD_ATTRS
=
$(STARTD_ATTRS), AverageCPUsUsage
##Check if filesystems are mounted
#STARTD_CRON_JOBLIST = NODEHEALTH
#STARTD_CRON_NODEHEALTH_EXECUTABLE = $(ETC)/git/scripts/healthcheck_cvmfs.sh
#STARTD_CRON_NODEHEALTH_MODE = Periodic
#STARTD_CRON_NODEHEALTH_PERIOD = 600s
# Only accept RemoteJobs which will finish in time
START
=
$(START) &&
\
( TARGET.Owner=?= "mschnepf" ) &&
\
( TARGET.RemoteJob =?= TRUE ) &&
\
( $(TimeToLive) > TARGET.RequestWalltime )
This diff is collapsed.
Click to expand it.
scripts/config_sites.json
+
26
−
1
View file @
5523bd9c
...
...
@@ -225,6 +225,31 @@
"name"
:
"config_sites.json"
}
]
},
"Singularity"
:
{
"config"
:
[
"0001.constants.cfg"
,
"0002.linux.cfg"
,
"0010.node_base.cfg"
,
"0012.node_slave.cfg"
,
"0015.logging.cfg"
,
"0021.security.cfg"
,
"0060.node_worker.cfg"
,
"0065.node_worker_remote.cfg"
,
"0083.partitionable_slots.cfg"
,
"0085.auto_shutdown.cfg"
,
"0088.ssh_to_job.cfg"
,
"0220.singularity_test.cfg"
],
"scripts"
:
[
{
"destination"
:
"/etc/condor/"
,
"name"
:
"config_pull.py"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"shutdown.sh"
}
]
}
}
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment