1RSL(5) GRAM5 Commands RSL(5)
2
6 rsl - GRAM5 RSL Attributes
7
9 arguments
10 The command line arguments for the executable. Use quotes, if a
11 space is required in a single argument.
12 count
14 The number of executions of the executable.
15 directory
17 Specifies the path of the directory the jobmanager will use as the
18 default directory for the requested job.
19 dry_run
21 If dryrun = yes then the jobmanager will not submit the job for
22 execution and will return success.
23 environment
25 The environment variables that will be defined for the executable
26 in addition to default set that is given to the job by the
27 jobmanager.
28 executable
30 The name of the executable file to run on the remote machine. If
31 the value is a GASS URL, the file is transferred to the remote gass
32 cache before executing the job and removed after the job has
33 terminated.
34 file_clean_up
36 Specifies a list of files which will be removed after the job is
37 completed.
38 file_stage_in
40 Specifies a list of ("remote URL" "local file") pairs which
41 indicate files to be staged to the nodes which will run the job.
42 file_stage_in_shared
44 Specifies a list of ("remote URL" "local file") pairs which
45 indicate files to be staged into the cache. A symlink from the
46 cache to the "local file" path will be made.
47 file_stage_out
49 Specifies a list of ("local file" "remote URL") pairs which
50 indicate files to be staged from the job to a GASS-compatible file
51 server.
52 gass_cache
54 Specifies location to override the GASS cache location.
55 gram_my_job
57 Obsolete and ignored.
58 host_count
60 Only applies to clusters of SMP computers, such as newer IBM SP
61 systems. Defines the number of nodes ("pizza boxes") to distribute
62 the "count" processes across.
63 job_type
65 This specifies how the jobmanager should start the job. Possible
66 values are single (even if the count > 1, only start 1 process or
67 thread), multiple (start count processes or threads), mpi (use the
68 appropriate method (e.g. mpirun) to start a program compiled with a
69 vendor-provided MPI library. Program is started with count nodes),
70 and condor (starts condor jobs in the "condor" universe.)
71 library_path
73 Specifies a list of paths to be appended to the system-specific
74 library path environment variables.
75 loglevel
77 Override the default log level for this job. The value of this
78 attribute consists of a combination of the strings FATAL, ERROR,
79 WARN, INFO, DEBUG, TRACE joined by the | character
80 logpattern
82 Override the default log path pattern for this job. The value of
83 this attribute is a string (potentially containing RSL
84 substitutions) that is evaluated to the path to write the log to.
85 If the resulting string contains the pattern $(DATE) (or any other
86 RSL substitution, it will be reevaluated at log time).
87 max_cpu_time
89 Explicitly set the maximum cputime for a single execution of the
90 executable. The units is in minutes. The value will go through an
91 atoi() conversion in order to get an integer. If the GRAM scheduler
92 cannot set cputime, then an error will be returned.
93 max_memory
95 Explicitly set the maximum amount of memory for a single execution
96 of the executable. The units is in Megabytes. The value will go
97 through an atoi() conversion in order to get an integer. If the
98 GRAM scheduler cannot set maxMemory, then an error will be
99 returned.
100 max_time
102 The maximum walltime or cputime for a single execution of the
103 executable. Walltime or cputime is selected by the GRAM scheduler
104 being interfaced. The units is in minutes. The value will go
105 through an atoi() conversion in order to get an integer.
106 max_wall_time
108 Explicitly set the maximum walltime for a single execution of the
109 executable. The units is in minutes. The value will go through an
110 atoi() conversion in order to get an integer. If the GRAM scheduler
111 cannot set walltime, then an error will be returned.
112 min_memory
114 Explicitly set the minimum amount of memory for a single execution
115 of the executable. The units is in Megabytes. The value will go
116 through an atoi() conversion in order to get an integer. If the
117 GRAM scheduler cannot set minMemory, then an error will be
118 returned.
119 project
121 Target the job to be allocated to a project account as defined by
122 the scheduler at the defined (remote) resource.
123 proxy_timeout
125 Obsolete and ignored. Now a job-manager-wide setting.
126 queue
128 Target the job to a queue (class) name as defined by the scheduler
129 at the defined (remote) resource.
130 remote_io_url
132 Writes the given value (a URL base string) to a file, and adds the
133 path to that file to the environment throught the
134 GLOBUS_REMOTE_IO_URL environment variable. If this is specified as
135 part of a job restart RSL, the job manager will update the file´s
136 contents. This is intended for jobs that want to access files via
137 GASS, but the URL of the GASS server has changed due to a GASS
138 server restart.
139 restart
141 Start a new job manager, but instead of submitting a new job, start
142 managing an existing job. The job manager will search for the job
143 state file created by the original job manager. If it finds the
144 file and successfully reads it, it will become the new manager of
145 the job, sending callbacks on status and streaming stdout/err if
146 appropriate. It will fail if it detects that the old jobmanager is
147 still alive (via a timestamp in the state file). If stdout or
148 stderr was being streamed over the network, new stdout and stderr
149 attributes can be specified in the restart RSL and the jobmanager
150 will stream to the new locations (useful when output is going to a
151 GASS server started by the client that´s listening on a dynamic
152 port, and the client was restarted). The new job manager will
153 return a new contact string that should be used to communicate with
154 it. If a jobmanager is restarted multiple times, any of the
155 previous contact strings can be given for the restart attribute.
156 rsl_substitution
158 Specifies a list of values which can be substituted into other rsl
159 attributes´ values through the $(SUBSTITUTION) mechanism.
160 save_state
162 Causes the jobmanager to save it´s job state information to a
163 persistent file on disk. If the job manager exits or is suspended,
164 the client can later start up a new job manager which can continue
165 monitoring the job.
166 savejobdescription
168 Save a copy of the job description to $HOME
169 scratch_dir
171 Specifies the location to create a scratch subdirectory in. A
172 SCRATCH_DIRECTORY RSL substitution will be filled with the name of
173 the directory which is created.
174 stderr
176 The name of the remote file to store the standard error from the
177 job. If the value is a GASS URL, the standard error from the job is
178 transferred dynamically during the execution of the job. There are
179 two accepted forms of this value. It can consist of a single
180 destination: stderr = URL, or a sequence of destinations: stderr =
181 (DESTINATION) (DESTINATION). In the latter case, the DESTINATION
182 may itself be a URL or a sequence of an x-gass-cache URL followed
183 by a cache tag."
184 stderr_position
186 Specifies where in the file remote standard error streaming should
187 be restarted from. Must be 0.
188 stdin
190 The name of the file to be used as standard input for the
191 executable on the remote machine. If the value is a GASS URL, the
192 file is transferred to the remote gass cache before executing the
193 job and removed after the job has terminated.
194 stdout
196 The name of the remote file to store the standard output from the
197 job. If the value is a GASS URL, the standard output from the job
198 is transferred dynamically during the execution of the job. There
199 are two accepted forms of this value. It can consist of a single
200 destination: stdout = URL, or a sequence of destinations: stdout =
201 (DESTINATION) (DESTINATION). In the latter case, the DESTINATION
202 may itself be a URL or a sequence of an x-gass-cache URL followed
203 by a cache tag.
204 stdout_position
206 Specifies where in the file remote output streaming should be
207 restarted from. Must be 0.
208 two_phase
210 Use a two-phase commit for job submission and completion. The job
211 manager will respond to the initial job request with a
212 WAITING_FOR_COMMIT error. It will then wait for a signal from the
213 client before doing the actual job submission. The integer supplied
214 is the number of seconds the job manager should wait before timing
215 out. If the job manager times out before receiving the commit
216 signal, or if a client issues a cancel signal, the job manager will
217 clean up the job´s files and exit, sending a callback with the job
218 status as GLOBUS_GRAM_PROTOCOL_JOB_STATE_FAILED. After the job
219 manager sends a DONE or FAILED callback, it will wait for a commit
220 signal from the client. If it receives one, it cleans up and exits
221 as usual. If it times out and save_state was enabled, it will leave
222 all of the job´s files in place and exit (assuming the client is
223 down and will attempt a job restart later). The timeoutvalue can be
224 extended via a signal. When one of the following errors occurs, the
225 job manager does not delete the job state file when it exits:
226 GLOBUS_GRAM_PROTOCOL_ERROR_COMMIT_TIMED_OUT,
227 GLOBUS_GRAM_PROTOCOL_ERROR_TTL_EXPIRED,
228 GLOBUS_GRAM_PROTOCOL_ERROR_JM_STOPPED,
229 GLOBUS_GRAM_PROTOCOL_ERROR_USER_PROXY_EXPIRED. In these cases, it
230 can not be restarted, so the job manager will not wait for the
231 commit signal after sending the FAILED callback
232 username
234 Verify that the job is running as this user.
235University of Chicago 03/08/2011 RSL(5)