[[condor:installation:configuration]]
Trace:
Show page
AdminRecent ChangesSitemap

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
Next revisionBoth sides next revision
condor:installation:configuration [2011/07/21 13:08] – updated configuration files garrettheath4condor:installation:configuration [2011/07/28 19:14] – garrettheath4
Line 1: Line 1:
-======Post-Install Configuration======+======Condor Configuration Files======
 This page contains copies of the configuration files we use on our system.  Use the table of contents to the right to jump to a specific configuration file. This page contains copies of the configuration files we use on our system.  Use the table of contents to the right to jump to a specific configuration file.
  
Line 159: Line 159:
 ## Join the W&L CS Pool (Orion) ## Join the W&L CS Pool (Orion)
 # Designate which machines are members of this pool. # Designate which machines are members of this pool.
-PoolMembers = john.cs.wlu.edu, carl.cs.wlu.edu, fred.cs.wlu.edu +# NOTE: In order to make it easy to automatically change this variable when 
-CondorUsers = condor@john.cs.wlu.edu/john.cs.wlu.edu, \ +#       running the installation script, the "PoolMembers" variable and its 
-              condor@carl.cs.wlu.edu/carl.cs.wlu.edu, \ +#       value should all be on one line (no line breaks). 
-              condor@fred.cs.wlu.edu/fred.cs.wlu.edu +PoolMembers = john.cs.wlu.edu, carl.cs.wlu.edu, fred.cs.wlu.edu, babbage.cs.wlu.edu
-AdminUsers = koller@$(CONDOR_HOST)/$(CONDOR_HOST) +
-RootUsers = root@john.cs.wlu.edu/john.cs.wlu.edu, \ +
-            root@carl.cs.wlu.edu/carl.cs.wlu.edu, \ +
-            root@fred.cs.wlu.edu/fred.cs.wlu.edu+
  
 ##  FLOCK_FROM defines the machines where you would like to grant ##  FLOCK_FROM defines the machines where you would like to grant
Line 201: Line 197:
 ##  Please see the administrator's manual for details on these ##  Please see the administrator's manual for details on these
 ##  settings, what they're for, and how to use them. ##  settings, what they're for, and how to use them.
 +
 +# Define fully-qualified usernames for users' right definitions
 +CondorUsers = condor@$(UID_DOMAIN)/john.cs.wlu.edu, \
 +              condor@$(UID_DOMAIN)/carl.cs.wlu.edu, \
 +              condor@$(UID_DOMAIN)/fred.cs.wlu.edu
 +AdminUsers = koller@$(UID_DOMAIN)/$(CONDOR_HOST)
 +RootUsers = root@$(UID_DOMAIN)/john.cs.wlu.edu, \
 +            root@$(UID_DOMAIN)/carl.cs.wlu.edu, \
 +            root@$(UID_DOMAIN)/fred.cs.wlu.edu
  
 ##  What machines have administrative rights for your pool?  This ##  What machines have administrative rights for your pool?  This
Line 233: Line 238:
 ##  you install, configure or debug your Condor installation. ##  you install, configure or debug your Condor installation.
 ##  It is important to have this defined. ##  It is important to have this defined.
-ALLOW_READ = $(AdminUsers), $(CondorUsers), *@cs.wlu.edu/*.cs.wlu.edu+ALLOW_READ = $(AdminUsers), $(CondorUsers), *@cs.wlu.edu/*.cs.wlu.edu, \ 
 +             *.cs.wisc.edu
 #ALLOW_READ = *.your.domain, *.cs.wisc.edu #ALLOW_READ = *.your.domain, *.cs.wisc.edu
 #DENY_READ = *.bad.subnet, bad-machine.your.domain, 144.77.88.* #DENY_READ = *.bad.subnet, bad-machine.your.domain, 144.77.88.*
Line 249: Line 255:
 ##  machines to your pool and is a serious security risk. ##  machines to your pool and is a serious security risk.
  
-ALLOW_WRITE = $(AdminUsers), $(CondorUsers), *@cs.wlu.edu/*.cs.wlu.edu+ALLOW_WRITE = $(AdminUsers), $(CondorUsers), \ 
 +              *@cs.wlu.edu/*.cs.wlu.edu
 #ALLOW_WRITE = *.your.domain, your-friend's-machine.other.domain #ALLOW_WRITE = *.your.domain, your-friend's-machine.other.domain
 #DENY_WRITE = bad-machine.your.domain #DENY_WRITE = bad-machine.your.domain
Line 266: Line 273:
 ##  Negotiator access.  Machines listed here are trusted central ##  Negotiator access.  Machines listed here are trusted central
 ##  managers.  You should normally not have to change this. ##  managers.  You should normally not have to change this.
-ALLOW_NEGOTIATOR = condor@$(CONDOR_HOST)/$(CONDOR_HOST)+#ALLOW_NEGOTIATOR = condor@$(CONDOR_HOST)/$(CONDOR_HOST) 
 +ALLOW_NEGOTIATOR = $(CONDOR_HOST)
 ##  Now, with flocking we need to let the SCHEDD trust the other  ##  Now, with flocking we need to let the SCHEDD trust the other 
 ##  negotiators we are flocking with as well.  You should normally  ##  negotiators we are flocking with as well.  You should normally 
Line 320: Line 328:
 # authentication methods when trying to communicate with the Condor # authentication methods when trying to communicate with the Condor
 # system daemons. # system daemons.
-SEC_CLIENT_AUTHENTICATION_METHODS = FS, PASSWORD+SEC_CLIENT_AUTHENTICATION = PREFERRED 
 +SEC_CLIENT_AUTHENTICATION_METHODS = PASSWORD, FS, FS_REMOTE 
 +SEC_CLIENT_INTEGRITY = OPTIONAL
  
 # A daemon will accept these forms of authentication when # A daemon will accept these forms of authentication when
 # communicating # communicating
-SEC_DEFAULT_AUTHENTICATION_METHODS = FS, PASSWORD+SEC_DEFAULT_AUTHENTICATION = PREFERRED 
 +SEC_DEFAULT_AUTHENTICATION_METHODS = PASSWORD, FS, FS_REMOTE 
 +SEC_DEFAULT_INTEGRITY = OPTIONAL
  
 # Password authentication # Password authentication
-SEC_PASSWORD_FILE = /var/lib/condor/pool_password +# Note: TILDE refers to the 'condor' user's home directory. 
-SEC_DAEMON_AUTHENTICATION = REQUIRED +SEC_PASSWORD_FILE = $(TILDE)/pool_password 
-SEC_DAEMON_INTEGRITY = REQUIRED +FS_REMOTE_DIR = $(CondorDir)/scratch 
-SEC_DAEMON_AUTHENTICATION_METHODS = PASSWORD + 
-SEC_NEGOTIATOR_AUTHENTICATION = REQUIRED +SEC_ADMINISTRATOR_AUTHENTICATION = REQUIRED 
-SEC_NEGOTIATOR_INTEGRITY = REQUIRED +SEC_CONFIG_AUTHENTICATION = REQUIRED 
-SEC_NEGOTIATOR_AUTHENTICATION_METHODS = PASSWORD+SEC_READ_AUTHENTICATION = OPTIONAL
  
  
Line 351: Line 363:
 ## uncomment the following setting and change the value to "True".  ## uncomment the following setting and change the value to "True". 
 ## Note: changing this requires a restart not just a reconfig. ## Note: changing this requires a restart not just a reconfig.
-ENABLE_RUNTIME_CONFIG = False+ENABLE_RUNTIME_CONFIG = True
  
 ## Do you want to allow condor_config_val -set to work at all? ## Do you want to allow condor_config_val -set to work at all?
Line 371: Line 383:
 ##  prior to version 6.3.3.  If you don't need this behavior, you ##  prior to version 6.3.3.  If you don't need this behavior, you
 ##  should leave this commented out. ##  should leave this commented out.
-#SETTABLE_ATTRS_CONFIG = *+SETTABLE_ATTRS_CONFIG = *
  
 ##  Attributes that can be set by hosts with "ADMINISTRATOR" ##  Attributes that can be set by hosts with "ADMINISTRATOR"
Line 551: Line 563:
 ##  default for this setting is False, since it is more secure this ##  default for this setting is False, since it is more secure this
 ##  way. ##  way.
-TRUST_UID_DOMAIN = True+TRUST_UID_DOMAIN = False
  
 ## If you would like to be informed in near real-time via condor_q when ## If you would like to be informed in near real-time via condor_q when
Line 631: Line 643:
 ## interface if one is available. If it cannot decide which of two interfaces ## interface if one is available. If it cannot decide which of two interfaces
 ## to choose from, it will pick the first one. ## to choose from, it will pick the first one.
-#NETWORK_INTERFACE = +#BIND_ALL_INTERFACES = True 
 +#NETWORK_INTERFACE = eth0
  
 ##-------------------------------------------------------------------- ##--------------------------------------------------------------------
Line 641: Line 654:
 ## ##
  
-ALL_DEBUG               =+#TODO 
 +ALL_DEBUG               = D_SECURITY
  
 MAX_COLLECTOR_LOG = 1000000 MAX_COLLECTOR_LOG = 1000000
Line 771: Line 785:
 ##  By default, RANK is always 0, meaning that all jobs have an equal ##  By default, RANK is always 0, meaning that all jobs have an equal
 ##  ranking. ##  ranking.
-RANK = $(GreedyUserRank)+RANK = $(GreedyUserRank) + ((TotalSlots > 1) * (SlotID / TotalSlots)) + ((KeyboardIdle > 15 * 60) * (5)) + (1 - LoadAvg)
  
  
Line 789: Line 803:
  
 # When should we only consider SUSPEND instead of PREEMPT? # When should we only consider SUSPEND instead of PREEMPT?
-WANT_SUSPEND = $(UWCS_WANT_SUSPEND)+#WANT_SUSPEND = $(UWCS_WANT_SUSPEND) 
 +WANT_SUSPEND = ( $(SmallJob) || $(KeyboardNotBusy) \ 
 +                            || $(IsVanilla) ) \ 
 +                          && ( $(SUSPEND) )
  
 # When should we preempt gracefully instead of hard-killing? # When should we preempt gracefully instead of hard-killing?
-WANT_VACATE = $(UWCS_WANT_VACATE)+#WANT_VACATE = $(UWCS_VACATE) 
 +WANT_VACATE = ( $(ActivationTimer) > 10 * $(MINUTE) \ 
 +                            || $(IsVanilla) )
  
 ##  When is this machine willing to start a job?  ##  When is this machine willing to start a job? 
-START = ($(UWCS_START) || $(IsGreedyUser))+#Note: START is redefined in the local config file of 'babbage.cs.wlu.edu' 
 +#START = ($(UWCS_START) || $(IsGreedyUser)) 
 +START = ( $(CPUIdle) || \ 
 +                            (State != "Unclaimed" && State != "Owner"))
  
 ##  When should a local universe job be allowed to start? ##  When should a local universe job be allowed to start?
Line 804: Line 826:
  
 ##  When to suspend a job? ##  When to suspend a job?
-SUSPEND = ($(UWCS_SUSPEND) && ($(IsGreedyUser) == False))+#SUSPEND = ($(UWCS_SUSPEND) && ($(IsGreedyUser) == False)) 
 +SUSPEND = ( (CpuBusyTime > 2 * $(MINUTE)) \ 
 +                            && $(ActivationTimer) > 90 )
  
 ##  When to resume a suspended job? ##  When to resume a suspended job?
-CONTINUE = ($(UWCS_CONTINUE) || $(IsGreedyUser))+#CONTINUE = ($(UWCS_CONTINUE) || $(IsGreedyUser)) 
 +CONTINUE = ( $(CPUIdle) && ($(ActivityTimer) > 5) )
  
 ##  When to nicely stop a job? ##  When to nicely stop a job?
 ##  (as opposed to killing it instantaneously) ##  (as opposed to killing it instantaneously)
-PREEMPT = ($(UWCS_PREEMPT) && ($(IsGreedyUser) == False))+#PREEMPT = ($(UWCS_PREEMPT) && ($(IsGreedyUser) == False)) 
 +PREEMPT = ( ((Activity == "Suspended") && \ 
 +                  ($(ActivityTimer) > $(MaxSuspendTime))) \ 
 + || (SUSPEND && (WANT_SUSPEND == False)) )
  
 ##  When to instantaneously kill a preempting job ##  When to instantaneously kill a preempting job
 ##  (e.g. if a job is in the pre-empting stage for too long) ##  (e.g. if a job is in the pre-empting stage for too long)
-KILL = ($(UWCS_KILL) && ($(IsGreedyUser) == False))+KILL = $(UWCS_KILL)
  
 PERIODIC_CHECKPOINT = $(UWCS_PERIODIC_CHECKPOINT) PERIODIC_CHECKPOINT = $(UWCS_PERIODIC_CHECKPOINT)
Line 1016: Line 1044:
 SPOOL = $(LOCAL_DIR)/spool SPOOL = $(LOCAL_DIR)/spool
 #EXECUTE = $(LOCAL_DIR)/execute #EXECUTE = $(LOCAL_DIR)/execute
-EXECUTE = /var/lib/condor/execute+# Note: TILDE refers to the 'condor' user's home directory. 
 +EXECUTE = $(TILDE)/execute
 BIN = $(RELEASE_DIR)/bin BIN = $(RELEASE_DIR)/bin
 LIB = $(RELEASE_DIR)/lib LIB = $(RELEASE_DIR)/lib
Line 1141: Line 1170:
 ## different port ranges for incoming and outgoing connections by ## different port ranges for incoming and outgoing connections by
 ## using IN_HIGHPORT/IN_LOWPORT and OUT_HIGHPORT/OUT_LOWPORT. ## using IN_HIGHPORT/IN_LOWPORT and OUT_HIGHPORT/OUT_LOWPORT.
-#HIGHPORT = 9700  +HIGHPORT = 9700  
-#LOWPORT = 9600+LOWPORT = 9600
  
 ##  If a daemon doens't respond for too long, do you want go generate ##  If a daemon doens't respond for too long, do you want go generate
Line 1534: Line 1563:
 ##  but you still want jobs to start right away on some of your SMP ##  but you still want jobs to start right away on some of your SMP
 ##  nodes, just increase this parameter. ##  nodes, just increase this parameter.
-#DISCONNECTED_KEYBOARD_IDLE_BOOST = 1200+DISCONNECTED_KEYBOARD_IDLE_BOOST = 1200
  
 ###### ######
Line 1596: Line 1625:
  
 ##  How often should the schedd send an update to the central manager? ##  How often should the schedd send an update to the central manager?
-SCHEDD_INTERVAL = 600 +SCHEDD_INTERVAL = 300 
  
 ##  How long should the schedd wait between spawning each shadow? ##  How long should the schedd wait between spawning each shadow?
Line 1603: Line 1632:
 ##  How many concurrent sub-processes should the schedd spawn to handle ##  How many concurrent sub-processes should the schedd spawn to handle
 ##  queries?  (Unix only) ##  queries?  (Unix only)
-SCHEDD_QUERY_WORKERS   = 5+SCHEDD_QUERY_WORKERS   = 10
  
 ##  How often should the schedd send a keep alive message to any ##  How often should the schedd send a keep alive message to any
Line 1699: Line 1728:
 ##  How often should the shadow update the job queue with job ##  How often should the shadow update the job queue with job
 ##  attributes that periodically change?  Specified in seconds. ##  attributes that periodically change?  Specified in seconds.
-SHADOW_QUEUE_UPDATE_INTERVAL = 15 * 60+SHADOW_QUEUE_UPDATE_INTERVAL = 10 * 60
  
 ##  Should the shadow wait to update certain job attributes for the ##  Should the shadow wait to update certain job attributes for the
Line 1731: Line 1760:
 ##  exit with an error.  Do you want the starter to just start up the ##  exit with an error.  Do you want the starter to just start up the
 ##  job with the specified UID, even if it's not in the passwd file? ##  job with the specified UID, even if it's not in the passwd file?
-#SOFT_UID_DOMAIN = FALSE+SOFT_UID_DOMAIN = FALSE
  
 ##  honor the run_as_owner option from the condor submit file. ##  honor the run_as_owner option from the condor submit file.
Line 1792: Line 1821:
 ##  site, uncomment these entries. ##  site, uncomment these entries.
 #APPEND_REQUIREMENTS = (expression to append job requirements) #APPEND_REQUIREMENTS = (expression to append job requirements)
-#APPEND_RANK = (expression to append job rank)+#APPEND_RANK = 
  
 ##  If you want expressions only appended for either standard or ##  If you want expressions only appended for either standard or
Line 2613: Line 2642:
 ## name node (see HDFS_SERVICES) then the specified port will be used ## name node (see HDFS_SERVICES) then the specified port will be used
 ## to run name node.  ## to run name node. 
-HDFS_NAMENODE = hdfs://example.com:9000 +#HDFS_NAMENODE = hdfs://example.com:9000 
-HDFS_NAMENODE_WEB = example.com:8000+#HDFS_NAMENODE_WEB = example.com:8000
  
-HDFS_BACKUPNODE = hdfs://example.com:50100 +#HDFS_BACKUPNODE = hdfs://example.com:50100 
-HDFS_BACKUPNODE_WEB = example.com:50105+#HDFS_BACKUPNODE_WEB = example.com:50105
  
 ## You need to pick one machine as name node by setting this parameter ## You need to pick one machine as name node by setting this parameter
condor/installation/configuration.txt · Last modified: 2011/08/19 18:52 by garrettheath4
Show pageOld revisions
Media ManagerBack to top
CC Attribution-Noncommercial-Share Alike 4.0 International
Driven by DokuWiki Recent changes RSS feed Valid CSS Valid XHTML 1.0