Submitting a Job
The condor_submit command takes a job description file as input and submits the job to HTCondor. In the submit description file, HTCondor finds everything it needs to know about the job. Items such as the name of the executable to run, the initial working directory, and command-line arguments to the program all go into the submit description file. condor_submit creates a job ClassAd based upon the information, and HTCondor works toward running the job.
It is easy to submit multiple runs of a program
to HTCondor with a single submit description file. To run the same
program many times with different input data sets, arrange the data files
accordingly so that each run reads its own input, and each run writes
its own output. Each individual run may have its own initial working
directory, files mapped for
command-line arguments, and shell environment.
The condor_submit manual page contains a complete and full description of how to use condor_submit. It also includes descriptions of all of the many commands that may be placed into a submit description file. In addition, the index lists entries for each command under the heading of Submit Commands.
Sample submit description files
In addition to the examples of submit description files given here, there are more in the condor_submit manual page.
Example 1 is one of the simplest submit description files possible. It queues the program myexe for execution somewhere in the pool. As this submit description file does not request a specific operating system to run on, HTCondor will use the default, which is to run the job on a machine which has the same architecture and operating system it was submitted from.
Before submitting a job to HTCondor, it is a good idea to test it first locally, by running it from a command shell. This example job might look like this when run from the shell prompt.
$ ./myexe SomeArgument
The corresponding submit description file might look like the following
# Example 1 # Simple HTCondor submit description file # Everything with a leading # is a comment executable = myexe arguments = SomeArgument output = outputfile error = errorfile log = myexe.log request_cpus = 1 request_memory = 1024M request_disk = 10240K should_transfer_files = yes queue
The standard output for this job will go to the file
outputfile, as specified by the
output command. Likewise,
the standard error output will go to
HTCondor will append events about the job to a log file with the
myexe.log. When the job
finishes, its exit conditions and resource usage will also be noted in the log file.
This file’s contents are an excellent way to figure out what happened to jobs.
HTCondor needs to know how many machine resources to allocate to this job.
request_ lines describe that this job should be allocated 1 cpu core, 1024
megabytes of memory and 10240 kilobytes of scratch disk space.
Finally, the queue statement tells HTCondor that you are done describing the job, and to send it to the queue for processing.
The submit description file for Example 2 queues 150
runs of program foo.
This job requires machines which have at least
4 GiB of physical memory, one cpu core and 16 Gb of scratch disk.
Each of the 150 runs of the program is given its own HTCondor process number,
starting with 0. $(Process) is expanded by HTCondor to the actual number
used by each instance of the job. So,
stderr will refer to
err.0 for the first run of the program,
err.1 for the second run of the program,
and so forth. A log file containing entries about when and where
HTCondor runs, transfer files, and terminates for all the 150
queued programs will be written into the single file
If there are 150 or more available slots in your pool, all 150 instances
might be run at the same time, otherwise, HTCondor will run as many as
it can concurrently.
Each instance of this program works on one input file. The name of this input file is passed to the program as the only argument. We prepare 150 copies of this input file in the current directory, and name them input_file.0, input_file.1 … up to input_file.149. Using transfer_input_files, we tell HTCondor which input file to send to each instance of the program.
# Example 2: Show off some fancy features, # including the use of pre-defined macros. executable = foo arguments = input_file.$(Process) request_cpus = 1 request_memory = 4096M request_disk = 16383K error = err.$(Process) output = out.$(Process) log = foo.log should_transfer_files = yes transfer_input_files = input_file.$(Process) # submit 150 instances of this job queue 150
Submitting many similar jobs with one queue command
A wide variety of job submissions can be specified with extra information to the queue submit command. This flexibility eliminates the need for a job wrapper or Perl script for many submissions.
The form of the queue command defines variables and expands values, identifying a set of jobs. Square brackets identify an optional item.
queue [<int expr> ]
queue [<int expr> ] [<varname> ] in [slice ] <list of items>
queue [<int expr> ] [<varname> ] matching [files | dirs ] [slice ] <list of items with file globbing>
queue [<int expr> ] [<list of varnames> ] from [slice ] <file name> | <list of items>
All optional items have defaults:
<int expr>is not specified, it defaults to the value 1.
<list of varnames>is not specified, it defaults to the single variable called
sliceis not specified, it defaults to all elements within the list. This is the Python slice
[::], with a step value of 1.
dirsis specified in a specification using the from key word, then both files and directories are considered when globbing.
The list of items uses syntax in one of two forms. One form is a comma
and/or space separated list; the items are placed on the same line as
the queue command. The second form separates items by placing each
list item on its own line, and delimits the list with parentheses. The
opening parenthesis goes on the same line as the queue command. The
closing parenthesis goes on its own line. The queue command
specified with the key word from will always use the second form of
this syntax. Example 3 below uses this second form of syntax. Finally,
the key word from accepts a shell command in place of file name,
followed by a pipe
| (example 4).
slice specifies a subset of the list of items using the
Python syntax for a slice. Negative step values are not permitted.
Here are a set of examples.
transfer_input_files = $(filename) arguments = -infile $(filename) queue filename matching files *.dat
The use of file globbing expands the list of items to be all files in
the current directory that end in
.dat. Only files, and not
directories are considered due to the specification of
job is queued for each file in the list of items. For this example,
assume that the three files
ending.dat form the list of items after expansion; macro
filename is assigned the value of one of these file names for each
job queued. That macro value is then substituted into the arguments
and transfer_input_files commands. The queue command expands
transfer_input_files = initial.dat arguments = -infile initial.dat queue transfer_input_files = middle.dat arguments = -infile middle.dat queue transfer_input_files = ending.dat arguments = -infile ending.dat queue
queue 1 input in A, B, C
input is set to each of the 3 items in the list, and one
job is queued for each. For this example the queue command expands
input = A queue input = B queue input = C queue
queue input, arguments from ( file1, -a -b 26 file2, -c -d 92 )
from form of the options, each of the two variables
specified is given a value from the list of items. For this example the
queue command expands to
input = file1 arguments = -a -b 26 queue input = file2 arguments = -c -d 92 queue
queue from seq 7 9 |
feeds the list of items to queue with the output of
seq 7 9:
item = 7 queue item = 8 queue item = 9 queue
Variables in the Submit Description File
There are automatic variables for use within the submit description file.
Each set of queued jobs from a specific user, submitted from a single submit host, sharing an executable have the same value of
$(ClusterId). The first cluster of jobs are assigned to cluster 0, and the value is incremented by one for each new cluster of jobs.
$(ClusterId)will have the same value as the job ClassAd attribute
Within a cluster of jobs, each takes on its own unique
$(ProcId)value. The first job has value 0.
$(ProcId)will have the same value as the job ClassAd attribute
When the machine is matched to this job for it to run on, any dollar-dollar expressions are looked up from the machine ad, and then expanded. This lets you put the value of some machine ad attribute into your job. For example, if you to pass the actual amount of memory a slot has provisioned as an argument to the job, you could add
arguments = --mem $$(Memory)
arguments = --mem $$(Memory)
or, if you wanted to put the name of the machine the job ran on into the output file name, you could add
output = output_file.$$(Name)
$$([ an_evaluated_classad_expression ])
This dollar-dollar-bracket syntax is useful when you need to perform some math on a value before passing it to your job. For example, if want to pass 90% of the allocated memory as an argument to your job, the submit file can have
arguments = --mem $$([ Memory * 0.9 ])
and when the job is matched to a machine, condor will evaluate this expression in the context of both the job and machine ad
The Architecture that HTCondor is running on, or the ARCH variable in the config file. Example might be X86_64.
These submit file macros are availle at submit time, and mimic the classad attributes of the same names.
The name of the submit_file as passed to the
The Unix epoch time submit was run. Note, this may be useful for naming output files.
These integer values are derived from the $(SUBMIT_TIME) macro above.
The default name of the variable when no
<varname>is provided in a queue command.
Represents an index within a list of items. When no slice is specified, the first
$(ItemIndex)is 0. When a slice is specified,
$(ItemIndex)is the index of the item within the original list.
$(Step)counts, starting at 0.
When a list of items is specified by placing each item on its own line in the submit description file,
$(Row)identifies which line the item is on. The first item (first line of the list) is
$(Row)0. The second item (second line of the list) is
$(Row)1. When a list of items are specified with all items on the same line,
$(Row)is the same as
Here is an example of a queue command for which the values of these automatic variables are identified.
This example queues six jobs.
queue 3 in (A, B)
$(Process)takes on the six values 0, 1, 2, 3, 4, and 5.
Because there is no specification for the
<varname>within this queue command, variable
$(Item)is defined. It has the value
Afor the first three jobs queued, and it has the value
Bfor the second three jobs queued.
$(Step)takes on the three values 0, 1, and 2 for the three jobs with
$(Item)=A, and it takes on the same three values 0, 1, and 2 for the three jobs with
$(ItemIndex)is 0 for all three jobs with
$(Item)=A, and it is 1 for all three jobs with
$(Row)has the same value as
$(ItemIndex)for this example.
Including Submit Commands Defined Elsewhere
Externally defined submit commands can be incorporated into the submit description file using the syntax
include : <what-to-include>
The <what-to-include> specification may specify a single file, where the
contents of the file will be incorporated into the submit description
file at the point within the file where the include is. Or,
<what-to-include> may cause a program to be executed, where the output
of the program is incorporated into the submit description file. The
specification of <what-to-include> has the bar character (
following the name of the program to be executed.
The include key word is case insensitive. There are no requirements for white space characters surrounding the colon character.
Included submit commands may contain further nested include specifications, which are also parsed, evaluated, and incorporated. Levels of nesting on included files are limited, such that infinite nesting is discovered and thwarted, while still permitting nesting.
Consider the example
include : ./list-infiles.sh |
In this example, the bar character at the end of the line causes the
list-infiles.sh to be invoked, and the output of the script
is parsed and incorporated into the submit description file. If this
bash script is in the PATH when submit is run, and contains
#!/bin/sh echo "transfer_input_files = `ls -m infiles/*.dat`" exit 0
then the output of this script has specified the set of input files to
transfer to the execute host. For example, if directory
contains the three files
C.dat, then the
transfer_input_files = infiles/A.dat, infiles/B.dat, infiles/C.dat
is incorporated into the submit description file.
Using Conditionals in the Submit Description File
Conditional if/else semantics are available in a limited form. The syntax:
if <simple condition> <statement> . . . <statement> else <statement> . . . <statement> endif
An else key word and statements are not required, such that simple if semantics are implemented. The <simple condition> does not permit compound conditions. It optionally contains the exclamation point character (!) to represent the not operation, followed by
the defined keyword followed by the name of a variable. If the variable is defined, the statement(s) are incorporated into the expanded input. If the variable is not defined, the statement(s) are not incorporated into the expanded input. As an example,
if defined MY_UNDEFINED_VARIABLE X = 12 else X = -1 endif
X = -1, when
MY_UNDEFINED_VARIABLEis not yet defined.
the version keyword, representing the version number of of the daemon or tool currently reading this conditional. This keyword is followed by an HTCondor version number. That version number can be of the form x.y.z or x.y. The version of the daemon or tool is compared to the specified version number. The comparison operators are
== for equality. Current version 8.2.3 is equal to 8.2.
>= to see if the current version number is greater than or equal to. Current version 8.2.3 is greater than 8.2.2, and current version 8.2.3 is greater than or equal to 8.2.
<= to see if the current version number is less than or equal to. Current version 8.2.0 is less than 8.2.2, and current version 8.2.3 is less than or equal to 8.2.
As an example,
if version >= 8.1.6 DO_X = True else DO_Y = True endif
results in defining
Trueif the current version of the daemon or tool reading this if statement is 8.1.6 or a more recent version.
True or yes or the value 1. The statement(s) are incorporated.
False or no or the value 0 The statement(s) are not incorporated.
$(<variable>) may be used where the immediately evaluated value is a simple boolean value. A value that evaluates to the empty string is considered False, otherwise a value that does not evaluate to a simple boolean value is a syntax error.
if <simple condition> <statement> . . . <statement> elif <simple condition> <statement> . . . <statement> endif
is the same as syntax
if <simple condition> <statement> . . . <statement> else if <simple condition> <statement> . . . <statement> endif endif
Here is an example use of a conditional in the submit description file.
A portion of the
sample.sub submit description file uses the if/else
syntax to define command line arguments in one of two ways:
if defined X arguments = -n $(X) else arguments = -n 1 -debug endif
X is defined on the condor_submit command line
$ condor_submit X=3 sample.sub
This command line incorporates the submit command
X = 3 into the
submission before parsing the submit description file. For this
submission, the command line arguments of the submitted job become
arguments = -n 3
If the job were instead submitted with the command line
$ condor_submit sample.sub
then the command line arguments of the submitted job become
arguments = -n 1 -debug
Function Macros in the Submit Description File
A set of predefined functions increase flexibility. Both submit description files and configuration files are read using the same parser, so these functions may be used in both submit description files and configuration files.
Case is significant in the function’s name, so use the same letter case as given in these definitions.
$CHOICE(index, item1, item2, ...)
An item within the list is returned. The list is represented by a parameter name, or the list items are the parameters. The
indexparameter determines which item. The first item in the list is at index 0. If the index is out of bounds for the list contents, an error occurs.
Evaluates to the value of environment variable
environment-variable-name. If there is no environment variable with that name, Evaluates to UNDEFINED unless the optional :default-value is used; in which case it evaluates to default-value. For example,
A = $ENV(HOME)
Ato the value of the
One or more of the lower case letters may be combined to form the function name and thus, its functionality. Each letter operates on the
filenamein its own way.
fconvert relative path to full path by prefixing the current working directory to it. This option works only in condor_submit files.
prefers to the entire directory portion of
filename, with a trailing slash or backslash character. Whether a slash or backslash is used depends on the platform of the machine. The slash will be recognized on Linux platforms; either a slash or backslash will be recognized on Windows platforms, and the parser will use the same character specified.
drefers to the last portion of the directory within the path, if specified. It will have a trailing slash or backslash, as appropriate to the platform of the machine. The slash will be recognized on Linux platforms; either a slash or backslash will be recognized on Windows platforms, and the parser will use the same character specified unless u or w is used. if b is used the trailing slash or backslash will be omitted.
uconvert path separators to Unix style slash characters
wconvert path separators to Windows style backslash characters
nrefers to the file name at the end of any path, but without any file name extension. As an example, the return value from
xrefers to a file name extension, with the associated period (
.). As an example, the return value from
bwhen combined with the d option, causes the trailing slash or backslash to be omitted. When combined with the x option, causes the leading period (
.) to be omitted.
qcauses the return value to be enclosed within quotes. Double quote marks are used unless a is also specified.
aWhen combined with the q option, causes the return value to be enclosed within single quotes.
$DIRNAME(filename) is the same as
$BASENAME(filename) is the same as
Expands, evaluates, and returns a string version of
format-specifierhas the same syntax as a C language or Perl format specifier. If no
format-specifieris specified, “%d” is used as the format specifier.
$RANDOM_CHOICE(choice1, choice2, choice3, ...)
A random choice of one of the parameters in the list of parameters is made. For example, if one of the integers 0-8 (inclusive) should be randomly chosen:
$RANDOM_INTEGER(min, max [, step])
A random integer within the range min and max, inclusive, is selected. The optional step parameter controls the stride within the range, and it defaults to the value 1. For example, to randomly chose an even integer in the range 0-8 (inclusive):
$RANDOM_INTEGER(0, 8, 2)
Expands, evaluates, and returns a string version of
item-to-convertfor a floating point type. The
format-specifieris a C language or Perl format specifier. If no
format-specifieris specified, “%16G” is used as a format specifier.
$SUBSTR(name, start-index, length)
Expands name and returns a substring of it. The first character of the string is at index 0. The first character of the substring is at index start-index. If the optional length is not specified, then the substring includes characters up to the end of the string. A negative value of start-index works back from the end of the string. A negative value of length eliminates use of characters from the end of the string. Here are some examples that all assume
Name = abcdef
$SUBSTR(Name, 0, -2)is
$SUBSTR(Name, 1, 3)is
$SUBSTR(Name, 4, -3)is the empty string, as there are no characters in the substring for this request.
Here are example uses of the function macros in a submit description file. Note that these are not complete submit description files, but only the portions that promote understanding of use cases of the function macros.
Generate a range of numerical values for a set of jobs, where values other than those given by $(Process) are desired.
MyIndex = $(Process) + 1 initial_dir = run-$INT(MyIndex,%04d)
Assuming that there are three jobs queued, such that $(Process) becomes
0, 1, and 2,
initial_dir will evaluate to the directories
This variation on Example 1 generates a file name extension which is a 3-digit integer value.
Values = $(Process) * 10 Extension = $INT(Values,%03d) input = X.$(Extension)
Assuming that there are four jobs queued, such that $(Process) becomes
0, 1, 2, and 3,
Extension will evaluate to 000, 010, 020, and 030,
leading to files defined for input of
This example uses both the file globbing of the queue command and a macro function to specify a job input file that is within a subdirectory on the submit host, but will be placed into a single, flat directory on the execute host.
arguments = $Fnx(FILE) transfer_input_files = $(FILE) queue FILE matching ( samplerun/*.dat )
Assume that two files that end in
within the directory
FILE expands to
samplerun/B.dat for the two jobs queued. The
input files transferred are
on the submit host. The
$Fnx() function macro expands to the
complete file name with any leading directory specification stripped,
such that the command line argument for one of the jobs will be
A.dat and the command line argument for the other job will be
About Requirements and Rank
rank commands in the submit description
file are powerful and flexible.
Using them effectively requires
care, and this section presents those details.
rank need to be specified as valid
HTCondor ClassAd expressions, however, default values are set by the
condor_submit program if these are not defined in the submit
description file. From the condor_submit manual page and the above
examples, you see that writing ClassAd expressions is intuitive,
especially if you are familiar with the programming language C. There
are some pretty nifty expressions you can write with ClassAds. A
complete description of ClassAds and their expressions can be found in
the HTCondor’s ClassAd Mechanism section.
All of the commands in the submit description file are case insensitive, except for the ClassAd attribute string values. ClassAd attribute names are case insensitive, but ClassAd string values are case preserving.
Note that the comparison operators (<, >, <=, >=, and ==) compare strings case insensitively. The special comparison operators =?= and =!= compare strings case sensitively.
A requirements or rank command in the submit description file may utilize attributes that appear in a machine or a job ClassAd. Within the submit description file (for a job) the prefix MY. (on a ClassAd attribute name) causes a reference to the job ClassAd attribute, and the prefix TARGET. causes a reference to a potential machine or matched machine ClassAd attribute.
The condor_status command displays statistics about machines within the pool. The -l option displays the machine ClassAd attributes for all machines in the HTCondor pool. The job ClassAds, if there are jobs in the queue, can be seen with the condor_q -l command. This shows all the defined attributes for current jobs in the queue.
A list of defined ClassAd attributes for job ClassAds is given in the Appendix on the Job ClassAd Attributes page. A list of defined ClassAd attributes for machine ClassAds is given in the Appendix on the Machine ClassAd Attributes page.
Rank Expression Examples
When considering the match between a job and a machine, rank is used to choose a match from among all machines that satisfy the job’s requirements and are available to the user, after accounting for the user’s priority and the machine’s rank of the job. The rank expressions, simple or complex, define a numerical value that expresses preferences.
Rank expression evaluates to one of three values. It can
be UNDEFINED, ERROR, or a floating point value. If
Rank evaluates to
a floating point value, the best match will be the one with the largest,
positive value. If no
Rank is given in the submit description file,
then HTCondor substitutes a default value of 0.0 when considering
machines to match. If the job’s
Rank of a given machine evaluates to
UNDEFINED or ERROR, this same value of 0.0 is used. Therefore, the
machine is still considered for a match, but has no ranking above any
A boolean expression evaluates to the numerical value of 1.0 if true, and 0.0 if false.
Rank expressions provide examples to follow.
For a job that desires the machine with the most available memory:
Rank = memory
For a job that prefers to run on a friend’s machine on Saturdays and Sundays:
Rank = ( (clockday == 0) || (clockday == 6) ) && (machine == "friend.cs.wisc.edu")
For a job that prefers to run on one of three specific machines:
Rank = (machine == "friend1.cs.wisc.edu") || (machine == "friend2.cs.wisc.edu") || (machine == "friend3.cs.wisc.edu")
For a job that wants the machine with the best floating point performance (on Linpack benchmarks):
Rank = kflops
This particular example highlights a difficulty with
evaluation as currently defined. While all machines have floating point
processing ability, not all machines will have the
defined. For machines where this attribute is not defined,
evaluate to the value UNDEFINED, and HTCondor will use a default rank of
the machine of 0.0. The
Rank attribute will only rank machines where
the attribute is defined. Therefore, the machine with the highest
floating point performance may not be the one given the highest rank.
So, it is wise when writing a
Rank expression to check if the
expression’s evaluation will lead to the expected resulting ranking of
machines. This can be accomplished using the condor_status command
with the -constraint argument. This allows the user to see a list of
machines that fit a constraint. To see which machines in the pool have
kflops defined, use
$ condor_status -constraint kflops
Alternatively, to see a list of machines where
kflops is not
$ condor_status -constraint "kflops=?=undefined"
For a job that prefers specific machines in a specific order:
Rank = ((machine == "friend1.cs.wisc.edu")*3) + ((machine == "friend2.cs.wisc.edu")*2) + (machine == "friend3.cs.wisc.edu")
If the machine being ranked is
friend1.cs.wisc.edu, then the
(machine == "friend1.cs.wisc.edu")
is true, and gives the value 1.0. The expressions
(machine == "friend2.cs.wisc.edu")
(machine == "friend3.cs.wisc.edu")
are false, and give the value 0.0. Therefore,
Rank evaluates to the
value 3.0. In this way, machine
friend1.cs.wisc.edu is ranked higher
ranked higher than machine
friend3.cs.wisc.edu, and all three of
these machines are ranked higher than others.
Jobs That Require Credentials
If the HTCondor pool administrator has configured the access point with one or more credential monitors, jobs submitted on that machine may automatically be provided with credentials and/or it may be possible for users to request and obtain credentials for their jobs.
Suppose the administrator has configured the access point such that users may obtain credentials from a storage service called “CloudBoxDrive.” A job that needs credentials from CloudBoxDrive should contain the submit command
use_oauth_services = cloudboxdrive
Upon submitting this job for the first time,
the user will be directed to a webpage hosted on the access point
which will guide the user through the process of obtaining a CloudBoxDrive credential.
The credential is then stored securely on the access point.
(Note: depending on which credential monitor is used, the original
job may have to be re-submitted at this point.)
(Also note that at no point is the user’s password stored on the access point.)
Once a credential is stored on the access point,
as long as it remains valid,
it is transferred securely to all subsequently submitted jobs that contain
use_oauth_services = cloudboxdrive.
When a job that contains credentials runs on an execute machine,
the job’s executable will have the environment variable
which points to the location of all of the credentials inside the job’s sandbox.
For credentials obtained via the
use_oauth_services submit file command,
the “access token” is stored under
in a JSON-encoded file
named with the name of the service provider and with the extension
For the “CloudBoxDrive” example,
the access token would be located in
The HTCondor file transfer mechanism has built-in plugins for using user-obtained credentials to transfer files from some specific storage providers, see File Transfer Using a URL.
Some credential providers may require the user to provide
a description of the permissions (often called “scopes”) a user needs for a specific credential.
Credential permission scoping is possible using the
submit file command.
For example, suppose our CloudBoxDrive service has a
and the documentation for the service said that users must specify a
in order to be able to read data out of
The submit file would need to contain
use_oauth_services = cloudboxdrive cloudboxdrive_oauth_permissions = read:/public
Some credential providers may also require the user to provide
the name of the resource (or “audience”) that a credential should allow access to.
Resource naming is done using the
<service name>_oauth_resource submit file command.
For example, if our CloudBoxDrive service has servers located at some universities
and the documentation says that we should pick one near us and specify it as the audience,
the submit file might look like
use_oauth_services = cloudboxdrive cloudboxdrive_oauth_permissions = read:/public cloudboxdrive_oauth_resource = https://cloudboxdrive.myuni.edu
It is possible for a single job to request and/or use credentials from multiple services
by listing each service in the
Suppose the nearby university has a SciTokens service that provides credentials to access the
and the HTCondor pool administrator has configured the access point to allow users to obtain credentials from this service,
and that a user has write access to the /foo directory on the storage machine.
A submit file that would result in a job that contains credentials
that can read from CloudBoxDrive and write to the local university storage might look like
use_oauth_services = cloudboxdrive, myuni cloudboxdrive_oauth_permissions = read:/public cloudboxdrive_oauth_resource = https://cloudboxdrive.myuni.edu myuni_oauth_permissions = write:/foo myuni_oauth_resource = https://localstorage.myuni.edu
A single job can also request multiple credentials from the same service provider
by affixing handles to the
<service>_oauth_permissions and (if necessary)
For example, if a user wants separate read and write credentials for CloudBoxDrive
use_oauth_services = cloudboxdrive cloudboxdrive_oauth_permissions_readpublic = read:/public cloudboxdrive_oauth_permissions_writeprivate = write:/private cloudboxdrive_oauth_resource_readpublic = https://cloudboxdrive.myuni.edu cloudboxdrive_oauth_resource_writeprivate = https://cloudboxdrive.myuni.edu
Submitting the above would result in a job with respective access tokens located in
Note that the permissions and resource settings for each handle (and for no handle) are stored separately from the job so multiple jobs from the same user running at the same time or for a period of time consecutively may not use a different set of permissions and resource settings for the same service and handle. If that is attempted, a new job submission will fail with instructions on how to resolve the conflict, but the safest thing is to choose a unique handle.
If a service provider does not require permissions or resources to be specified,
a user can still request multiple credentials by affixing handles to
<service>_oauth_permissions commands with empty values
use_oauth_services = cloudboxdrive cloudboxdrive_oauth_permissions_personal = cloudboxdrive_oauth_permissions_public =
When the Vault credential monitor is configured, the service name may optionally be split into two parts with an underscore between them, where the first part is the issuer and the second part is the role. In this example the issuer is “dune” and the role is “production”, both as configured by the administrator of the Vault server:
use_oauth_services = dune_production
Vault does not require permissions or resources to be
set, but they may be set to reduce the default permissions or restrict
the resources that may use the credential. The full service name
including an underscore may be used in an
oauth_resource. Avoid using handles that might be confused as
role names. For example, the following will result in a conflict
between two credentials called
use_oauth_services = dune, dune_production dune_oauth_permissions_production = dune_production_oauth_permissions =
Jobs That Require GPUs
HTCondor has built-in support for detecting machines with GPUs, and matching jobs that need GPUs to machines that have them. If your job needs a GPU, you’ll first need to tell HTCondor how many GPUs each job needs with the submit command:
request_GPUs = <n>
<n> is replaced by the integer quantity of GPUs required for
the job. For example, a job that needs 1 GPU uses
request_GPUs = 1
Because there are different capabilities among GPUs, your job might need to further qualify which GPU is required. The submit command require_gpus does this. For example, to request a CUDA GPU whose CUDA Capability is at least 8, add the following to your submit file:
request_GPUs = 1 require_gpus = Capability >= 8.0
To see which CUDA capabilities are available in your HTCondor pool, you can run the command
$ condor_status -af Name GPUS_Capability
To see which GPU devices HTCondor has detected on your pool, you can run the command
$ condor_status -af Name GPUS_DeviceName
Access to GPU resources by an HTCondor job needs special configuration of the machines that offer GPUs. Details of how to set up the configuration are in the Policy Configuration for Execution Points and for Access Points section.
An interactive job is a Condor job that is provisioned and scheduled like any other vanilla universe Condor job onto an execute machine within the pool. The result of a running interactive job is a shell prompt issued on the execute machine where the job runs. The user that submitted the interactive job may then use the shell as desired, perhaps to interactively run an instance of what is to become a Condor job. This might aid in checking that the set up and execution environment are correct, or it might provide information on the RAM or disk space needed. This job (shell) continues until the user logs out or any other policy implementation causes the job to stop running. A useful feature of the interactive job is that the users and jobs are accounted for within Condor’s scheduling and priority system.
Neither the submit nor the execute host for interactive jobs may be on Windows platforms.
The current working directory of the shell will be the initial working directory of the running job. The shell type will be the default for the user that submits the job. At the shell prompt, X11 forwarding is enabled.
Each interactive job will have a job ClassAd attribute of
InteractiveJob = True
Submission of an interactive job specifies the option -interactive on the condor_submit command line.
A submit description file may be specified for this interactive job. Within this submit description file, a specification of these 5 commands will be either ignored or altered:
universe . The interactive job is a vanilla universe job.
queue <n>. In this case the value of <n> is ignored; exactly one interactive job is queued.
The submit description file may specify anything else needed for the interactive job, such as files to transfer.
If no submit description file is specified for the job, a default one is
utilized as identified by the value of the configuration variable
Here are examples of situations where interactive jobs may be of benefit.
An application that cannot be batch processed might be run as an interactive job. Where input or output cannot be captured in a file and the executable may not be modified, the interactive nature of the job may still be run on a pool machine, and within the purview of Condor.
A pool machine with specialized hardware that requires interactive handling can be scheduled with an interactive job that utilizes the hardware.
The debugging and set up of complex jobs or environments may benefit from an interactive session. This interactive session provides the opportunity to run scripts or applications, and as errors are identified, they can be corrected on the spot.
Development may have an interactive nature, and proceed more quickly when done on a pool machine. It may also be that the development platforms required reside within Condor’s purview as execute hosts.
Submitting Lots of Jobs
When submitting a lot of jobs with a single submit file, you can dramatically speed up submission and reduce the load on the condor_schedd by submitting the jobs as a late materialization job factory.
A submission of this form sends a single ClassAd, called the Cluster ad, to the condor_schedd, as
well as instructions to create the individual jobs as variations on that Cluster ad. These instructions
are sent as a submit digest and optional itemdata. The submit digest is the submit file stripped down
to just the statements that vary between jobs. The itemdata is the arguments to the
when the arguments are more than just a count of jobs.
The condor_schedd will use the submit digest and the itemdata to create the individual job ClassAds when they are needed. Materialization is controlled by two values stored in the Cluster classad, and by optional limits configured in the condor_schedd.
max_idle limit specifies the maximum number of non-running jobs that should be materialized in the
condor_schedd at any one time. One or more jobs will materialize whenever a job enters the Run state
and the number of non-running jobs that are still in the condor_schedd is less than this limit. This
limit is stored in the Cluster ad in the JobMaterializeMaxIdle attribute.
max_materialize limit specifies an overall limit on the number of jobs that can be materialized in
the condor_schedd at any one time. One or more jobs will materialize when a job leaves the condor_schedd
and the number of materialized jobs remaining is less than this limit. This limit is stored in the Cluster
ad in the JobMaterializeLimit attribute.
Late materialization can be used as a way for a user to submit millions of jobs without hitting the
MAX_JOBS_PER_OWNER or MAX_JOBS_PER_SUBMISSION limits in the condor_schedd, since
the condor_schedd will enforce these limits by applying them to the
values specified in the Cluster ad.
To give an example, the following submit file:
executable = foo arguments = input_file.$(Process) request_cpus = 1 request_memory = 4096M request_disk = 16383K error = err.$(Process) output = out.$(Process) log = foo.log should_transfer_files = yes transfer_input_files = input_file.$(Process) # submit as a factory with an idle jobs limit max_idle = 100 # submit 15,000 instances of this job queue 15*1000
When submitted as a late materialization factory, the submit digest for this factory will contain only the submit statements that vary between jobs, and the collapsed queue statement like this:
arguments = input_file.$(Process) error = err.$(Process) output = out.$(Process) transfer_input_files = input_file.$(Process) queue 15000
Materialization log events
When a Late Materialization job factory is submitted to the condor_schedd, a
Cluster submitted event
will be written to the UserLog of the Cluster ad. This will be the same log file used by the first job
materialized by the factory. To avoid confusion,
it is recommended that you use the same log file for all jobs in the factory.
When the Late Materialization job factory is removed from the condor_schedd, a
Cluster removed event
will be written to the UserLog of the Cluster ad. This event will indicate how many jobs were materialized
before the factory was removed.
If Late Materialization of jobs is paused due to an error in materialization or because condor_hold
was used to hold the cluster id, a
Job Materialization Paused event will be written to the UserLog of the
Cluster ad. This event will indicate the reason for the pause.
condor_release is used to release the the cluster id of a Late Materialization job factory,
and materialization was paused because of a previous use of condor_hold, a
Job Materialization Resumed
event will be written to the UserLog of the Cluster ad.
Currently, not all features of condor_submit will work with late materialization. The following limitations apply:
Only a single
Queuestatement is allowed, lines from the submit file after the first
Queuestatement will be ignored.
$RANDOM_CHOICEmacro functions will expand at submit time to produce the Cluster ad, but these macro functions will not be included in the submit digest and so will have the same value for all jobs.
Spooling of input files does not work with late materialization.
Displaying the Factory
condor_q can be use to show late materialization job factories in the condor_schedd by
> condor_q -factory -- Schedd: submit.example.org : <192.168.101.101:9618?... @ 12/01/20 13:35:00 ID OWNER SUBMITTED LIMIT PRESNT RUN IDLE HOLD NEXTID MODE DIGEST 77. bob 12/01 13:30 15000 130 30 80 20 1230 /var/lib/condor/spool/77/condor_submit.77.digest
The factory above shows that 30 jobs are currently running,
80 are idle, 20 are held and that the next job to materialize will
77.1230. The total of Idle + Held jobs is 100,
which is equal to the
max_idle value specified in the submit file.
The path to the submit digest file is shown. This file is used to reload the factory
when the condor_schedd is restarted. If the factory is unable to materialize jobs
because of an error, the
MODE field will show
Errs to indicate
there is a problem.
Errs indicates a problem reloading the factory,
indicates a problem materializing jobs.
In case of a factory problem, use
condor_q -factory -long to see the the factory information
Removing a Factory
The Late materialization job factory will be remove from the schedd automatically once all of the jobs have materialized and completed. To remove the factory without first completing all of the jobs use condor_rm with the ClusterId of the factory as the argument.
Editing a Factory
The submit digest for a Late Materialization job factory cannot be changed after submission, but the Cluster ad for the factory can be edited using condor_qedit. Any condor_qedit command that has the ClusterId as a edit target will edit all currently materialized jobs, as well as editing the Cluster ad so that all jobs that materialize in the future will also be edited.