Mailing List Archives Authenticated access	UW Madison Computer Sciences Department Computer Systems Lab

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[HTCondor-users] preempting many jobs before maxjobretirementtime is reached

Date: Thu, 21 May 2020 19:02:45 +0200
From: Carsten Aulbert <carsten.aulbert@xxxxxxxxxx>
Subject: [HTCondor-users] preempting many jobs before maxjobretirementtime is reached

Hi,

I hope everyone has fully recovered from the HTCondor week :)

Unfortunately, this problem only came up after yesterday's Q&A session
finished, thus asking it in the usual space.

In our new configuration (config summary at the end) we try to enforce
preemption based on priority (and maybe other factors in the future),
but right now, jobs are killed earlier than what MaxJobRetirementTime is
set to and we currently have no idea what is causing this?

The lines copied below seem to indicate networking problems, but I do
not see any hints in the switches or on the nodes - especially the
submit machine is quite busy but it should not overwhelm it (system is
mostly idle but with about 5% wait, 10% system and mostly pegasus
monitor processes in D state).

The service should have enough file handles:

condor.service - Condor Distributed High-Throughput-Computing
Loaded: loaded (/lib/systemd/system/condor.service; enabled; vendor
preset: enabled)
Drop-In: /etc/systemd/system.control/condor.service.d
ââ50-TasksMax.conf
/etc/systemd/system/condor.service.d
ââensure-run-dir.conf, limits.conf
Active: active (running) since Thu 2020-05-21 08:30:44 UTC; 8h ago
Process: 2440231 ExecStartPre=/bin/mkdir --parent --mode=0755
/run/condor (code=exited, status=0/SUCCESS)
Process: 2440232 ExecStartPre=/bin/chown condor:condor /run/condor
(code=exited, status=0/SUCCESS)
Main PID: 2440233 (condor_master)
Status: "All daemons are responding"
Tasks: 34255 (limit: 131071)
Memory: 76.7G
CGroup: /system.slice/condor.service

Thus, I would be grateful for any hint where to look.

Cheers

Carsten

In the SchedLog we see these worrisome lines:

05/21/20 16:43:26 (2061018.0) (1646287): **** condor_shadow
(condor_SHADOW) pid 1646287 EXITING WITH STATUS 108
05/21/20 16:43:26 (2061021.0) (1646309): **** condor_shadow
(condor_SHADOW) pid 1646309 EXITING WITH STATUS 108
05/21/20 16:43:26 (2006521.0) (1624977): File transfer completed
successfully.
05/21/20 16:43:26 (2001479.0) (1610449): File transfer completed
successfully.
05/21/20 16:43:26 (2013966.0) (1637250): **** condor_shadow
(condor_SHADOW) pid 1637250 EXITING WITH STATUS 108
05/21/20 16:43:26 (2054171.0) (1339388): condor_read(): Socket closed
abnormally when trying to read 21 bytes from schedd at
<10.20.30.17:14867>, errno=104 Connection reset by peer
05/21/20 16:43:26 (2054171.0) (1339388): SetEffectiveOwner(ahnitz)
failed with errno=110: Connection timed out.
05/21/20 16:43:26 (2052187.0) (1313342): condor_read(): Socket closed
abnormally when trying to read 21 bytes from schedd at
<10.20.30.17:14867>, errno=104 Connection reset by peer
05/21/20 16:43:26 (2052187.0) (1313342): SetEffectiveOwner(ahnitz)
failed with errno=110: Connection timed out.
05/21/20 16:43:26 (2031965.0) (1283878): condor_read(): Socket closed
abnormally when trying to read 21 bytes from schedd at
<10.20.30.17:14867>, errno=104 Connection reset by peer
05/21/20 16:43:26 (2031965.0) (1283878): SetEffectiveOwner(ahnitz)
failed with errno=110: Connection timed out.
05/21/20 16:43:26 (2051766.0) (1308655): condor_read(): Socket closed
abnormally when trying to read 21 bytes from schedd at
<10.20.30.17:14867>, errno=104 Connection reset by peer
05/21/20 16:43:26 (2039452.0) (1293958): condor_read(): Socket closed
abnormally when trying to read 21 bytes from schedd at
<10.20.30.17:14867>, errno=104 Connection reset by peer

In StarterLogs these lines

05/21/20 15:24:32 (pid:48660) Create_Process succeeded, pid=48814
05/21/20 15:24:40 (pid:48660) Failed to open '.update.ad' to read update
ad: No such file or directory (2).
05/21/20 15:24:40 (pid:48660) Failed to open '.update.ad' to read update
ad: No such file or directory (2).
05/21/20 16:45:03 (pid:48660) Connection to shadow may be lost, will
test by sending whoami request.
05/21/20 16:45:03 (pid:48660) condor_write(): Socket closed when trying
to write 37 bytes to <10.20.30.17:12065>, fd is 9
05/21/20 16:45:03 (pid:48660) Buf::write(): condor_write() failed
05/21/20 16:45:03 (pid:48660) i/o error result is 0, errno is 0
05/21/20 16:45:03 (pid:48660) Lost connection to shadow, waiting 2400
secs for reconnect
05/21/20 16:45:27 (pid:48660) Got SIGTERM. Performing graceful shutdown.
05/21/20 16:45:27 (pid:48660) ShutdownGraceful all jobs.
05/21/20 16:45:28 (pid:48660) Process exited, pid=48814, status=143
05/21/20 16:45:28 (pid:48660) Returning from CStarter::JobReaper()

matching lines from StartLog:
05/21/20 15:11:00 slot1_64: New machine resource of type -1 allocated
05/21/20 15:11:01 slot1_64: Request accepted.
05/21/20 15:11:01 slot1_64: Remote owner is ahnitz@xxxxxxxxxxx
05/21/20 15:11:01 slot1_64: State change: claiming protocol successful
05/21/20 15:11:01 slot1_64: Changing state: Owner -> Claimed
05/21/20 15:21:01 slot1_64: Response problem from schedd
<10.20.30.17:9618?addrs=10.20.30.17-9618&noUDP&sock=2440233_7254_3> on
ALIVE job -1.-1.
05/21/20 15:21:06 slot1_64: Response problem from schedd
<10.20.30.17:9618?addrs=10.20.30.17-9618&noUDP&sock=2440233_7254_3> on
ALIVE job -1.-1.
05/21/20 15:21:11 slot1_64: Couldn't send ALIVE to schedd at
<10.20.30.17:9618?addrs=10.20.30.17-9618&noUDP&sock=2440233_7254_3>
05/21/20 15:23:16 slot1_64: Got activate_claim request from shadow
(10.20.30.17)
05/21/20 15:23:16 slot1_64: Remote job ID is 2049848.0
05/21/20 15:23:16 slot1_64: Got universe "VANILLA" (5) from request classad
05/21/20 15:23:16 slot1_64: State change: claim-activation protocol
successful
05/21/20 15:23:16 slot1_64: Changing activity: Idle -> Busy
05/21/20 15:23:19 slot1_64: Failed to open
'/local/condor/execute/dir_48660/.update.ad.tmp' for writing update ad:
No such file or directory (2).
05/21/20 15:24:01 slot1_64: Failed to open
'/local/condor/execute/dir_48660/.update.ad.tmp' for writing update ad:
No such file or directory (2).
05/21/20 15:24:04 slot1_64: Failed to open
'/local/condor/execute/dir_48660/.update.ad.tmp' for writing update ad:
No such file or directory (2).
05/21/20 15:24:04 slot1_64: Failed to open
'/local/condor/execute/dir_48660/.update.ad.tmp' for writing update ad:
No such file or directory (2).
05/21/20 16:45:27 slot1_64: Got ACTIVATE_CLAIM while in Claimed/Busy
state, ignoring.
05/21/20 16:45:27 slot1_64: Called deactivate_claim()
05/21/20 16:45:27 slot1_64: Changing state and activity: Claimed/Busy ->
Preempting/Vacating
05/21/20 16:45:38 slot1_64: Got RELEASE_CLAIM while in Preempting state,
ignoring.


-- 
Dr. Carsten Aulbert, Max Planck Institute for Gravitational Physics,
CallinstraÃe 38, 30167 Hannover, Germany
Phone: +49 511 762 17185

# condor_config_val $CondorVersion: 8.8.9 May 06 2020 BuildID: Debian-8.8.9-1 PackageID: 8.8.9-1 Debian-8.8.9-1 $

#
# from /etc/condor/condor_config
#
RELEASE_DIR = /usr
LOCAL_DIR = /local/condor
LOCAL_CONFIG_FILE = /etc/condor/condor_config.local
REQUIRE_LOCAL_CONFIG_FILE = false
LOCAL_CONFIG_DIR = /etc/condor/config.d
LOCAL_CONFIG_DIR_EXCLUDE_REGEXP = ^((\..*)|(.*~)|(#.*)|(.*\.rpmsave)|(.*\.rpmnew))$
ALLOW_NEGOTIATOR_SCHEDD = $(ALLOW_NEGOTIATOR) $(FLOCK_NEGOTIATOR_HOSTS)
ALLOW_READ_COLLECTOR = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_READ_STARTD = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_WRITE_COLLECTOR = $(ALLOW_WRITE) $(FLOCK_FROM)
ALLOW_WRITE_STARTD = $(ALLOW_WRITE) $(FLOCK_FROM)
RUN = /run/condor
LIB = $(RELEASE_DIR)/lib/condor
INCLUDE = $(RELEASE_DIR)/include/condor
LIBEXEC = $(RELEASE_DIR)/lib/condor/libexec
SHARE = $(RELEASE_DIR)/share/condor
PROCD_ADDRESS = $(RUN)/procd_pipe

#
# from /etc/condor/config.d/01_generic
#
LOG = /var/log/condor
MAX_MASTER_LOG = 100000000
MAX_CKPT_SERVER_LOG = 100000000
MAX_STARTD_LOG = 500000000
MAX_COLLECTOR_LOG = 500000000
MAX_NEGOTIATOR_LOG = 500000000
MAX_HAD_LOG = 100000000
MAX_SCHEDD_LOG = 500000000
MAX_SHADOW_LOG = 500000000
MAX_MATCH_LOG = 500000000
MAX_HISTORY_LOG = 1000000000
MAX_HISTORY_ROTATIONS = 7
ROTATE_HISTORY_DAILY = True
HISTORY_HELPER_MAX_HISTORY = 999999999
SEC_PASSWORD_FILE = /var/lib/condor/passwd
SEC_DAEMON_AUTHENTICATION = REQUIRED
SEC_DAEMON_INTEGRITY = REQUIRED
SEC_DAEMON_AUTHENTICATION_METHODS = PASSWORD
SEC_NEGOTIATOR_AUTHENTICATION = REQUIRED
SEC_NEGOTIATOR_INTEGRITY = REQUIRED
SEC_NEGOTIATOR_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_CLIENT_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_ADMINISTRATOR_AUTHENTICATION = REQUIRED
SEC_ADMINISTRATOR_AUTHENTICATION_METHODS = PASSWORD
SEC_DEFAULT_AUTHENTICATION_METHODS = PASSWORD, FS
ADMINISTRATORS = condor@xxxxxxxxxxx/*.atlas.local
CENTRAL_MANAGER = condorhub.atlas.local
ALL_SCHEDD = condor*.atlas.local
ALL_STARTD = a*.atlas.local
ALL_NODES = $(CENTRAL_MANAGER), $(ALL_SCHEDD), $(ALL_STARTD)
COLLECTOR_NAME = MPI-GRAPHY-AEI-Hannover
CONDOR_IDS = 666.666
COLLECTOR_HOST = $(CENTRAL_MANAGER)
NEGOTIATOR_HOST = $(CENTRAL_MANAGER)
DEFAULT_DOMAIN_NAME = atlas.local
UID_DOMAIN = atlas.local
FILESYSTEM_DOMAIN = atlas.local
LOCK = /var/lock/
CONDOR_ADMIN = atlas_admin@xxxxxxxxxx
MAIL = /usr/bin/mail
LOCAL_CONDOR_SCRIPTS = /usr/share/condor

#
# from /etc/condor/config.d/10_SCHEDD
#
NETWORK_INTERFACE = 10.20.30.17
DAEMON_LIST = MASTER SCHEDD
ALLOW_ADMINISTRATOR = $(ADMINISTRATORS)
ALLOW_ADVERTISE_MASTER = $(ADMINISTRATORS)
ALLOW_ADVERTISE_SCHEDD = $(ADMINISTRATORS)
ALLOW_CLIENT = $(CENTRAL_MANAGER), $(ALL_SCHEDD), $(ALL_STARTD)
ALLOW_CONFIG = $(ADMINISTRATORS)
ALLOW_DAEMON = $(ADMINISTRATORS)
ALLOW_NEGOTIATOR = $(ADMINISTRATORS), $(CENTRAL_MANAGER)
ALLOW_OWNER = $(ADMINISTRATORS)
ALLOW_READ = $(ADMINISTRATORS), $(FULL_HOSTNAME)
ALLOW_WRITE = *
MAX_JOBS_RUNNING = 100000

#
# from /etc/condor/config.d/20_FLOCKING
#
FLOCK_TO = condorhubinteractive.atlas.local

#
# from /etc/condor/config.d/99_TRANSFORM
#
JOB_TRANSFORM_NAMES = TagJob
JOB_TRANSFORM_TagJob = [
Eval_set_AccountingGroup = join(".", split(toLower(AcctGroup), ".")[1], AcctGroupUser);
Eval_set_AcctGroup = toLower(AcctGroup);
]
SCHEDD_CLASSAD_USER_MAP_NAMES =  ValidSearchTags ValidSearchUsers
CLASSAD_USER_MAPFILE_ValidSearchTags = /etc/condor/accounting/valid_tags
CLASSAD_USER_MAPFILE_ValidSearchUsers = /etc/condor/accounting/valid_users
SUBMIT_REQUIREMENT_NAMES =  ValidateSearchTag ValidateSearchUser
SUBMIT_REQUIREMENT_ValidateSearchTag = JobUniverse == 7 || userMap("ValidSearchTags", AcctGroup) isnt undefined
SUBMIT_REQUIREMENT_ValidateSearchTag_REASON = strcat("Invalid value for search tag: ", AcctGroup ?: "<undefined>")
SUBMIT_REQUIREMENT_ValidateSearchUser = Debug(JobUniverse == 7 || userMap("ValidSearchUsers", Owner, AcctGroupUser) is AcctGroupUser || userMap("ValidSearchUsers", Owner) is undefined && Owner =?= AcctGroupUser)
SUBMIT_REQUIREMENT_ValidateSearchUser_REASON = strcat("Invalid value for search user: ", AcctGroupUser ?: "<undefined>", "\n", "       Valid values are: ",userMap("ValidSearchUsers", Owner))

# condor_config_val $CondorVersion: 8.8.7 Dec 26 2019 BuildID: Debian-8.8.7-1 PackageID: 8.8.7-1 Debian-8.8.7-1 $

#
# from /etc/condor/condor_config
#
RELEASE_DIR = /usr
LOCAL_DIR = /local/condor
LOCAL_CONFIG_FILE = /etc/condor/condor_config.local
REQUIRE_LOCAL_CONFIG_FILE = false
LOCAL_CONFIG_DIR = /etc/condor/config.d
LOCAL_CONFIG_DIR_EXCLUDE_REGEXP = ^((\..*)|(.*~)|(#.*)|(.*\.rpmsave)|(.*\.rpmnew))$
ALLOW_NEGOTIATOR_SCHEDD = $(ALLOW_NEGOTIATOR) $(FLOCK_NEGOTIATOR_HOSTS)
ALLOW_READ_COLLECTOR = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_READ_STARTD = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_WRITE_COLLECTOR = $(ALLOW_WRITE) $(FLOCK_FROM)
ALLOW_WRITE_STARTD = $(ALLOW_WRITE) $(FLOCK_FROM)
RUN = /run/condor
LIB = $(RELEASE_DIR)/lib/condor
INCLUDE = $(RELEASE_DIR)/include/condor
LIBEXEC = $(RELEASE_DIR)/lib/condor/libexec
SHARE = $(RELEASE_DIR)/share/condor
PROCD_ADDRESS = $(RUN)/procd_pipe

#
# from /etc/condor/config.d/01_generic
#
LOG = /var/log/condor
MAX_MASTER_LOG = 100000000
MAX_CKPT_SERVER_LOG = 100000000
MAX_STARTD_LOG = 500000000
MAX_COLLECTOR_LOG = 500000000
MAX_NEGOTIATOR_LOG = 500000000
MAX_HAD_LOG = 100000000
MAX_SCHEDD_LOG = 500000000
MAX_SHADOW_LOG = 500000000
MAX_MATCH_LOG = 500000000
MAX_HISTORY_LOG = 1000000000
MAX_HISTORY_ROTATIONS = 7
ROTATE_HISTORY_DAILY = True
HISTORY_HELPER_MAX_HISTORY = 999999999
SEC_PASSWORD_FILE = /var/lib/condor/passwd
SEC_DAEMON_AUTHENTICATION = REQUIRED
SEC_DAEMON_INTEGRITY = REQUIRED
SEC_DAEMON_AUTHENTICATION_METHODS = PASSWORD
SEC_NEGOTIATOR_AUTHENTICATION = REQUIRED
SEC_NEGOTIATOR_INTEGRITY = REQUIRED
SEC_NEGOTIATOR_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_CLIENT_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_ADMINISTRATOR_AUTHENTICATION = REQUIRED
SEC_ADMINISTRATOR_AUTHENTICATION_METHODS = PASSWORD
SEC_DEFAULT_AUTHENTICATION_METHODS = PASSWORD, FS
ADMINISTRATORS = condor@xxxxxxxxxxx/*.atlas.local
CENTRAL_MANAGER = condorhub.atlas.local
ALL_SCHEDD = condor*.atlas.local
ALL_STARTD = a*.atlas.local
ALL_NODES = $(CENTRAL_MANAGER), $(ALL_SCHEDD), $(ALL_STARTD)
COLLECTOR_NAME = MPI-GRAPHY-AEI-Hannover
CONDOR_IDS = 666.666
COLLECTOR_HOST = $(CENTRAL_MANAGER)
NEGOTIATOR_HOST = $(CENTRAL_MANAGER)
DEFAULT_DOMAIN_NAME = atlas.local
UID_DOMAIN = atlas.local
FILESYSTEM_DOMAIN = atlas.local
LOCK = /var/lock/
CONDOR_ADMIN = atlas_admin@xxxxxxxxxx
MAIL = /usr/bin/mail
LOCAL_CONDOR_SCRIPTS = /usr/share/condor

#
# from /etc/condor/config.d/10_CENTRALMANAGER
#
NETWORK_INTERFACE = 10.20.40.190
DAEMON_LIST = MASTER COLLECTOR NEGOTIATOR
ALLOW_ADMINISTRATOR = $(ADMINISTRATORS), $(FULL_HOSTNAME)
ALLOW_ADVERTISE_MASTER = $(ADMINISTRATORS), $(ALL_NODES)
ALLOW_CLIENT = *
ALLOW_CONFIG = $(ADMINISTRATORS)
ALLOW_DAEMON = $(ADMINISTRATORS), $(ALL_NODES)
ALLOW_NEGOTIATOR = $(ADMINISTRATORS), $(CENTRAL_MANAGER)
ALLOW_OWNER = $(ADMINISTRATORS), $(FULL_HOSTNAME)
ALLOW_READ = *
ALLOW_WRITE = *
NewUserBetterPrio = ( RemoteUserPrio > SubmitterUserPrio * 1.2 )
IsGPUJob = ( Target.RequestGPUs > 0 && My.RequestGPUs =?= 0 )
PREEMPTION_REQUIREMENTS = $(NewUserBetterPrio)
ALLOW_PSLOT_PREEMPTION = True

# condor_config_val $CondorVersion: 8.8.9 May 06 2020 BuildID: Debian-8.8.9-1 PackageID: 8.8.9-1 Debian-8.8.9-1 $

#
# from /etc/condor/condor_config
#
RELEASE_DIR = /usr
LOCAL_DIR = /local/condor
LOCAL_CONFIG_FILE = /etc/condor/condor_config.local
REQUIRE_LOCAL_CONFIG_FILE = false
LOCAL_CONFIG_DIR = /etc/condor/config.d
LOCAL_CONFIG_DIR_EXCLUDE_REGEXP = ^((\..*)|(.*~)|(#.*)|(.*\.rpmsave)|(.*\.rpmnew))$
ALLOW_NEGOTIATOR = $(CONDOR_HOST)
ALLOW_NEGOTIATOR_SCHEDD = $(ALLOW_NEGOTIATOR) $(FLOCK_NEGOTIATOR_HOSTS)
ALLOW_OWNER = $(FULL_HOSTNAME) $(IPV4_ADDRESS) $(IPV6_ADDRESS)
ALLOW_READ = *
ALLOW_READ_COLLECTOR = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_READ_STARTD = $(ALLOW_READ) $(FLOCK_FROM)
ALLOW_WRITE_COLLECTOR = $(ALLOW_WRITE) $(FLOCK_FROM)
ALLOW_WRITE_STARTD = $(ALLOW_WRITE) $(FLOCK_FROM)
ALLOW_WRITE = 10.0.0.0/9
RUN = /run/condor
LIB = $(RELEASE_DIR)/lib/condor
INCLUDE = $(RELEASE_DIR)/include/condor
LIBEXEC = $(RELEASE_DIR)/lib/condor/libexec
SHARE = $(RELEASE_DIR)/share/condor
PROCD_ADDRESS = $(RUN)/procd_pipe

#
# from /etc/condor/config.d/01_generic
#
LOG = /var/log/condor
MAX_MASTER_LOG = 100000000
MAX_CKPT_SERVER_LOG = 100000000
MAX_STARTD_LOG = 500000000
MAX_COLLECTOR_LOG = 500000000
MAX_NEGOTIATOR_LOG = 500000000
MAX_HAD_LOG = 100000000
MAX_SCHEDD_LOG = 500000000
MAX_SHADOW_LOG = 500000000
MAX_MATCH_LOG = 500000000
MAX_HISTORY_LOG = 1000000000
MAX_HISTORY_ROTATIONS = 7
ROTATE_HISTORY_DAILY = True
HISTORY_HELPER_MAX_HISTORY = 999999999
SEC_PASSWORD_FILE = /var/lib/condor/passwd
SEC_DAEMON_AUTHENTICATION = REQUIRED
SEC_DAEMON_INTEGRITY = REQUIRED
SEC_DAEMON_AUTHENTICATION_METHODS = PASSWORD
SEC_NEGOTIATOR_AUTHENTICATION = REQUIRED
SEC_NEGOTIATOR_INTEGRITY = REQUIRED
SEC_NEGOTIATOR_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_CLIENT_AUTHENTICATION_METHODS = FS, PASSWORD
SEC_ADMINISTRATOR_AUTHENTICATION = REQUIRED
SEC_ADMINISTRATOR_AUTHENTICATION_METHODS = PASSWORD
SEC_DEFAULT_AUTHENTICATION_METHODS = PASSWORD, FS
ADMINISTRATORS = condor@xxxxxxxxxxx/*.atlas.local
CENTRAL_MANAGER = condorhub.atlas.local
ALL_SCHEDD = condor*.atlas.local
ALL_STARTD = a*.atlas.local
ALL_NODES = $(CENTRAL_MANAGER), $(ALL_SCHEDD), $(ALL_STARTD)
COLLECTOR_NAME = MPI-GRAPHY-AEI-Hannover
CONDOR_IDS = 666.666
COLLECTOR_HOST = $(CENTRAL_MANAGER)
NEGOTIATOR_HOST = $(CENTRAL_MANAGER)
DEFAULT_DOMAIN_NAME = atlas.local
UID_DOMAIN = atlas.local
FILESYSTEM_DOMAIN = atlas.local
LOCK = /var/lock/
CONDOR_ADMIN = atlas_admin@xxxxxxxxxx
MAIL = /usr/bin/mail
LOCAL_CONDOR_SCRIPTS = /usr/share/condor

#
# from /etc/condor/config.d/10_EXECUTE
#
NETWORK_INTERFACE = 10.10.71.1
DAEMON_LIST = MASTER STARTD
MAXJOBRETIREMENTTIME = (10 * $(HOUR))
START = True && (MaxRunTimeHours * $(HOUR) <= MAXJOBRETIREMENTTIME)
NUM_SLOTS_TYPE_1 = 1
SLOT_TYPE_1 = cpus=100%, ram=501913, swap=0%
SLOT_TYPE_1_PARTITIONABLE = True
LOCALINFO = $(LOCAL_CONDOR_SCRIPTS)/lscpu_localuser_info
STARTD_CRON_JOBLIST =  LOCALINFO
STARTD_CRON_LOCALINFO_EXECUTABLE = $(LOCALINFO)
STARTD_CRON_LOCALINFO_PERIOD = 300
MASTER_NEW_BINARY_RESTART = peaceful

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Follow-Ups:
- Re: [HTCondor-users] preempting many jobs before maxjobretirementtime is reached
  - From: Bockelman, Brian

Prev by Date: Re: [HTCondor-users] execute hosts advertise loopback address
Next by Date: Re: [HTCondor-users] preempting many jobs before maxjobretirementtime is reached
Previous by thread: Re: [HTCondor-users] execute hosts advertise loopback address
Next by thread: Re: [HTCondor-users] preempting many jobs before maxjobretirementtime is reached
Index(es):
- Date
- Thread

Mailing List Archives

Authenticated access

[HTCondor-users] preempting many jobs before maxjobretirementtime is reached