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
10eb88f1
Commit
10eb88f1
authored
8 months ago
by
Sebastian Brommer
Browse files
Options
Downloads
Patches
Plain Diff
add nemo2 config
parent
e0e2bcc1
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
0212.node_worker_bwforcluster_nemo2.cfg
+110
-0
110 additions, 0 deletions
0212.node_worker_bwforcluster_nemo2.cfg
scripts/config_sites.json
+43
-0
43 additions, 0 deletions
scripts/config_sites.json
with
153 additions
and
0 deletions
0212.node_worker_bwforcluster_nemo2.cfg
0 → 100644
+
110
−
0
View file @
10eb88f1
# bwForCluster OpenStack Worker Setup
# -----------------------------------
# Cloud attributes
STARTD.CloudSite
=
"BWFORCLUSTER_NEMO2"
# STARTD_DEBUG = D_ANY
# START_DEBUG = D_ANY
# STARTER_DEBUG = D_FULLDEBUG
# MASTER_DEBUG = D_ANY
# Machine/Job environment
# Use TardisDroneUuid to identify drones in OBS
# TardisDroneUuid = "$ENV(TardisDroneUuid)"
# STARTD_NAME = $ENV(TardisDroneUuid)
# Matchmaking advertisement
STARTD.ProvidesIO
=
TRUE
STARTD.ProvidesCpu
=
TRUE
# Only allow token based authentication
SEC_DEFAULT_AUTHENTICATION_METHODS
=
PASSWORD, IDTOKENS, SCITOKENS
SEC_TOKEN_DIRECTORY
=
$(ETC)/tokens.d/
SEC_TOKEN_SYSTEM_DIRECTORY
=
$(ETC)/tokens.d/
DomainNemo
=
login2.nemo.privat
ALLOW_DEAMON
=
$(ALLOW_DEAMON), condor@$(DomainNemo), condor_pool@$(DomainNemo)
ALLOW_WRITE
=
$(ALLOW_WRITE), condor@$(DomainNemo), condor_pool@$(DomainNemo)
ALLOW_ADMINISTRATOR
=
$(ALLOW_ADMINISTRATOR),
\
condor@$(UID_DOMAIN)/$(IP_ADDRESS)
# Network settings
# PRIVATE_NETWORK_NAME = cloud.ruf.uni-freiburg.de
DEFAULT_DOMAIN_NAME
=
$(DomainNemo)
UID_DOMAIN
=
$(DomainNemo)
# 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
=
86400
# Read/set MachineMaxWalltime & MachineStarttime (set by cloud-init)
include
command
into
$(ETC)/walltime.cfg
:
python3
$(ETC)/meta_walltime_moab.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) )
# Set number of CPUs
NUM_CPUS
=
1
# Machine/Job environment
# Read /etc/environment, for example for http proxy
STARTD.JOB_INHERITS_STARTER_ENVIRONMENT
=
TRUE
include
command
into
$(ETC)/environment.cfg
:
$(ETC)/get_machine_env.sh
# 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
SINGULARITY_JOB
=
TRUE
SINGULARITY
=
/usr/bin/apptainer
SINGULARITY_HOOK_PREPARE_JOB
=
$(ETC)/git/scripts/prepare_singularity.py
SINGULARITY_IMAGE_EXPR
=
IfThenElse(TARGET.ContainerImage=?=undefined, "/cvmfs/unpacked.cern.ch/registry.hub.docker.com/mschnepf/slc7-condocker:latest", TARGET.ContainerImage)
SINGULARITY_RUN_TEST_BEFORE_JOB
=
FALSE
SINGULARITY_TARGET_DIR
=
/srv
MOUNT_UNDER_SCRATCH
=
# The container needs to use the execute directory as HOME directory
SINGULARITY_HOME
=
/srv
SINGULARITY_EXTRA_ARGUMENTS
=
"--userns --env SINGULARITY_BIND= --env APPTAINER_BIND= --env APPTAINER_BINDPATH="
# Only accept RemoteJobs which will finish in time
START
=
$(START) &&
\
( TARGET.RemoteJob =?= TRUE ) &&
\
( TARGET.Owner == "sbrommer" ) &&
\
( $(TimeToLive) > TARGET.RequestWalltime )
This diff is collapsed.
Click to expand it.
scripts/config_sites.json
+
43
−
0
View file @
10eb88f1
...
...
@@ -42,6 +42,49 @@
}
]
},
"BwForClusterNemo2"
:
{
"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"
,
"0082.slot_users.cfg"
,
"0083.partitionable_slots.cfg"
,
"0085.auto_shutdown.cfg"
,
"0088.ssh_to_job.cfg"
,
"0212.node_worker_bwforcluster_nemo2.cfg"
],
"scripts"
:
[
{
"destination"
:
"/etc/condor/"
,
"name"
:
"config_pull.py"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"meta_walltime_moab.py"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"get_machine_env.sh"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"collector_address.py"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"config_sites.json"
},
{
"destination"
:
"/etc/condor/"
,
"name"
:
"shutdown.sh"
}
]
},
"Collector"
:
{
"config"
:
[
"0001.constants.cfg"
,
...
...
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