1SUBMIT(1) Grid Engine User Commands SUBMIT(1)
2
6 qsub - submit a batch job to Grid Engine.
7 qsh - submit an interactive X-windows session to Grid Engine.
9 qlogin - submit an interactive login session to Grid Engine.
11 qrsh - submit an interactive rsh session to Grid Engine.
13 qalter - modify a pending batch job of Grid Engine.
15 qresub - submit a copy of an existing Grid Engine job.
17
19 qsub [ options ] [ command | -- [ command_args ]]
20 qsh [ options ] [ -- xterm_args ]
22 qlogin [ options ]
24 qrsh [ options ] [ command [ command_args ]]
26 qalter [ options ] wc_job_range_list [ -- [ command_args ]]
28 qalter [ options ] -u user_list | -uall [ -- [ command_args ]]
30 qresub [ options ] job_id_list
32
34 qsub submits batch jobs to the Grid Engine queuing system. Grid Engine
35 supports single- and multiple-node jobs. Command can be a path to a
36 binary or a script (see -b below) which contains the commands to be run
37 by the job using a shell (for example, sh(1) or csh(1)). Arguments to
38 the command are given as command_args to qsub . If command is handled
39 as a script then it is possible to embed flags in the script. If the
40 first two characters of a script line either match '#$' or are equal to
41 the prefix string defined with the -C option described below, the line
42 is parsed for embedded command flags.
43 qsh submits an interactive X-windows session to Grid Engine. An
45 xterm(1) is brought up from the executing machine with the display
46 directed either to the X-server indicated by the DISPLAY environment
47 variable or as specified with the -display qsh option. Interactive jobs
48 are not spooled if no resource is available to execute them. They are
49 either dispatched to a suitable machine for execution immediately or
50 the user submitting the job is notified by qsh that appropriate
51 resources to execute the job are not available. xterm_args are passed
52 to the xterm(1) executable. Note, however, that the -e and -ls xterm
53 options do not work with qsh .
54 qlogin is similar to qsh in that it submits an interactive job to the
56 queueing system. It does not open an xterm(1) window on the X display,
57 but uses the current terminal for user I/O. Usually, qlogin establishes
58 a telnet(1) connection with the remote host, using standard client- and
59 server-side commands. These commands can be configured with the qlo‐
60 gin_daemon and qlogin_command in the global and local configuration
61 settings of sge_conf(5).
62 On the server-side, an Grid Engine telnetd is used if the parameter is
64 not set. Otherwise, it should be set to something like
65 /usr/sbin/in.telnetd. On the client side, an Grid Engine telnet is
66 used it the parameter is not set. Otherwise, it should be set to some‐
67 thing like /usr/bin/telnet.
68 The client side command is automatically parameterized with the remote
70 host name and port number to which to connect, resulting in an invoca‐
71 tion like
72 /usr/bin/telnet my_exec_host 2442
74 for example. qlogin is invoked exactly like qsh and its jobs can only
76 run on INTERACTIVE queues. qlogin jobs can only be used if the
77 sge_execd(8) is running under the root account.
78 qrsh is similar to qlogin in that it submits an interactive job to the
80 queuing system. It uses the current terminal for user I/O. Usually,
81 qrsh establishes a rsh(1) connection with the remote host. If no com‐
82 mand is given to qrsh, an rlogin(1) session is established. The
83 server-side commands used can be configured with the rsh_daemon and
84 rlogin_daemon parameters in the global and local configuration settings
85 of sge_conf(5). An Grid Engine rshd or rlogind is used if the parame‐
86 ters are not set. If the parameters are set, they should be set to
87 something like /usr/sbin/in.rshd or /usr/sbin/in.rlogind. On the
88 client-side, the rsh_command and rlogin_command parameters can be set
89 in the global and local configuration settings of sge_conf(5). If they
90 are not set, special Grid Engine rsh(1) and rlogin(1) binaries deliv‐
91 ered with Grid Engine are used. Use the cluster configuration parame‐
92 ters to integrate mechanisms like ssh or the rsh(1) and rlogin(1)
93 facilities supplied with the operating system.
94 qrsh jobs can only run in INTERACTIVE queues unless the option -now no
96 is used (see below). They can also only be run, if the sge_execd(8) is
97 running under the root account.
98 qrsh provides an additional useful feature for integrating with inter‐
100 active tools providing a specific command shell. If the environment
101 variable QRSH_WRAPPER is set when qrsh is invoked, the command inter‐
102 preter pointed to by QRSH_WRAPPER will be executed to run qrsh commands
103 instead of the users login shell or any shell specified in the qrsh
104 command-line. The options -cwd, -v, -V, and -display only apply to
105 batch jobs.
106 qalter can be used to change the attributes of pending jobs. For array
108 jobs with a mix of running and pending tasks (see the -t option below),
109 modification with qalter only affects the pending tasks. qalter can
110 change most of the characteristics of a job (see the corresponding
111 statements in the OPTIONS section below), including those which were
112 defined as embedded flags in the script file (see above). Some submit
113 options, such as the job script, cannot be changed with qalter.
114 qresub allows the user to create jobs as copies of existing pending or
116 running jobs. The copied jobs will have exactly the same attributes as
117 the ones from which they were copied, except with a new job ID. The
118 only modification to the copied jobs supported by qresub is assignment
119 of a hold state with the -h option. This option can be used to first
120 copy a job and then change its attributes via qalter.
121 Only a user with manager privileges can use qresub on jobs submitted by
123 another user. Regular users can only use qresub on their own jobs.
124 For qsub, qsh, qrsh, and qlogin the administrator and the user may
126 define default request files (see sge_request(5)) which can contain any
127 of the options described below. If an option in a default request file
128 is understood by qsub and qlogin but not by qsh the option is silently
129 ignored if qsh is invoked. Thus you can maintain shared default request
130 files for both qsub and qsh.
131 A cluster-wide default request file may be placed under
133 $SGE_ROOT/$SGE_CELL/common/sge_request. User private default request
134 files are processed under the locations $HOME/.sge_request and
135 $cwd/.sge_request. The working directory local default request file has
136 the highest precedence, then the home directory located file and then
137 the cluster global file. The option arguments, the embedded script
138 flags and the options in the default request files are processed in the
139 following order:
140 left to right in the script line,
142 left to right in the default request files,
143 from top to bottom of the script file (qsub only),
144 from top to bottom of default request files,
145 from left to right of the command line.
146 In other words, the command line can be used to override the embedded
148 flags and the default request settings. The embedded flags, however,
149 will override the default settings.
150 Note, that the -clear option can be used to discard any previous set‐
152 tings at any time in a default request file, in the embedded script
153 flags, or in a command-line option. It is, however, not available with
154 qalter.
155 The options described below can be requested either hard or soft. By
157 default, all requests are considered hard until the -soft option (see
158 below) is encountered. The hard/soft status remains in effect until its
159 counterpart is encountered again. If all the hard requests for a job
160 cannot be met, the job will not be scheduled. Jobs which cannot be run
161 at the present time remain spooled.
162
164 -@ optionfile
165 Forces qsub, qrsh, qsh, or qlogin to use the options contained
166 in optionfile. The indicated file may contain all valid options.
167 Comment lines must start with a "#" sign.
168 -a date_time
170 Available for qsub and qalter only.
171 Defines or redefines the time and date at which a job is eligi‐
173 ble for execution. Date_time conforms to [[CC]]YY]MMDDhhmm[.SS],
174 where:
175 CC denotes the century in 2 digits.
177 YY denotes the year in 2 digits.
178 MM denotes the month in 2 digits.
179 DD denotes the day in 2 digits.
180 hh denotes the hour in 2 digits.
181 mm denotes the minute in 2 digits.
182 ss denotes the seconds in 2 digits (default 00).
183 If any of the optional date fields is omitted, the corresponding
185 value of the current date is assumed. If CC is not specified, a
186 YY of < 70 means 20YY.
187 Usage of this option may cause unexpected results if the clocks
188 of the hosts in the Grid Engine pool are out of sync. Also, the
189 proper behavior of this option very much depends on the correct
190 setting of the appropriate timezone, e.g. in the TZ environment
191 variable (see date(1) for details), when the Grid Engine daemons
192 sge_qmaster(8) and sge_execd(8) are invoked.
193 qalter allows changing this option even while the job executes.
195 The modified parameter will only be in effect after a restart or
196 migration of the job, however.
197 -ac variable[=value],...
199 Available for qsub, qsh, qrsh, qlogin and qalter only.
200 Adds the given name/value pair(s) to the job's context. Value
202 may be omitted. Grid Engine appends the given argument to the
203 list of context variables for the job. Multiple -ac, -dc, and
204 -sc options may be given. The order is important here.
205 qalter allows changing this option even while the job executes.
207 -A account_string
209 Available for qsub, qsh, qrsh, qlogin and qalter only.
210 Identifies the account to which the resource consumption of the
212 job should be charged. The account_string may be any arbitrary
213 ASCII alphanumeric string but may not contain "\n", "\t", "\r",
214 "/", ":", "@", "\", "*", or "?". In the absence of this param‐
215 eter Grid Engine will place the default account string "sge" in
216 the accounting record of the job.
217 qalter allows changing this option even while the job executes.
219 -b y[es]|n[o]
221 Available for qsub, qrsh only. qalter does not allow changing
222 this option. This option cannot be embedded in the script file
223 itself.
224 Gives the user the possibility to indicate explicitly whether
226 command should be treated as binary or script. If the value of
227 -b is 'y', then command may be a binary or script. The command
228 might not be accessible from the submission host. Nothing
229 except the path of the command will be transferred from the sub‐
230 mission host to the execution host. Path aliasing will be
231 applied to the path of command before command will be executed.
232 If the value of -b is 'n' then command needs to be a script and
234 it will be handled as script. The script file has to be accessi‐
235 ble by the submission host. It will be transferred to the execu‐
236 tion host. qsub/qrsh will search directive prefixes within
237 script.
238 qsub will implicitly use -b n whereas qrsh will apply the -b y
240 option if nothing else is specified.
241 -c occasion_specifier
243 Available for qsub and qalter only.
244 Defines or redefines whether the job should be checkpointed, and
246 if so, under what circumstances. The specification of the check‐
247 pointing occasions with this option overwrites the definitions
248 of the when parameter in the checkpointing environment (see
249 checkpoint(5)) referenced by the qsub -ckpt switch. Possible
250 values for occasion_specifier are
251 n no checkpoint is performed.
253 s checkpoint when batch server is shut down.
254 m checkpoint at minimum CPU interval.
255 x checkpoint when job gets suspended.
256 <interval> checkpoint in the specified time interval.
257 The minimum CPU interval is defined in the queue configuration
259 (see queue_conf(5) for details). <interval> has to be specified
260 in the format hh:mm:ss. The maximum of <interval> and the
261 queue's minimum CPU interval is used if <interval> is specified.
262 This is done to ensure that a machine is not overloaded by
263 checkpoints being generated too frequently.
264 -ckpt ckpt_name
266 Available for qsub and qalter only.
267 Selects the checkpointing environment (see checkpoint(5)) to be
269 used for checkpointing the job. Also declares the job to be a
270 checkpointing job.
271 -clear Available for qsub, qsh, qrsh, and qlogin only.
273 Causes all elements of the job to be reset to the initial
275 default status prior to applying any modifications (if any)
276 appearing in this specific command.
277 -cwd Available for qsub, qsh, qrsh and qalter only.
279 Execute the job from the current working directory. This switch
281 will activate Grid Engine's path aliasing facility, if the cor‐
282 responding configuration files are present (see sge_aliases(5)).
283 In the case of qalter, the previous definition of the current
285 working directory will be overwritten if qalter is executed from
286 a different directory than the preceding qsub or qalter.
287 qalter allows changing this option even while the job executes.
289 The modified parameter will only be in effect after a restart or
290 migration of the job, however.
291 -C prefix_string
293 Available for qsub and qrsh with script submission (-b n).
294 Prefix_string defines the prefix that declares a directive in
296 the job's command. The prefix is not a job attribute, but
297 affects the behavior of qsub and qrsh. If prefix is a null
298 string, the command will not be scanned for embedded directives.
299 The directive prefix consists of two ASCII characters which,
300 when appearing in the first two bytes of a script line, indicate
301 that what follows is an Grid Engine command. The default is
302 "#$".
303 The user should be aware that changing the first delimiting
304 character can produce unforeseen side effects. If the script
305 file contains anything other than a "#" character in the first
306 byte position of the line, the shell processor for the job will
307 reject the line and may exit the job prematurely.
308 If the -C option is present in the script file, it is ignored.
309 -dc variable,...
311 Available for qsub, qsh, qrsh, qlogin and qalter only.
312 Removes the given variable(s) from the job's context. Multiple
314 -ac, -dc, and -sc options may be given. The order is important.
315 qalter allows changing this option even while the job executes.
317 -display display_specifier
319 Available for qsh only.
320 Directs xterm(1) to use display_specifier in order to contact
322 the X server. The display_specifier has to contain the hostname
323 part of the display name (e.g. myhost:1). Local display names
324 (e.g. :0) cannot be used in grid environments. Values set with
325 the -display option overwrite settings from the submission envi‐
326 ronment and from -v command line options.
327 -dl date_time
329 Available for qsub, qsh, qrsh, qlogin and qalter only.
330 Specifies the deadline initiation time in [[CC]YY]MMDDhhmm[.SS]
332 format (see -a option above). The deadline initiation time is
333 the time at which a deadline job has to reach top priority to be
334 able to complete within a given deadline. Before the deadline
335 initiation time the priority of a deadline job will be raised
336 steadily until it reaches the maximum as configured by the Grid
337 Engine administrator.
338 This option is applicable only for users allowed to submit dead‐
339 line jobs.
340 -e [[hostname]:]path,...
342 Available for qsub, qsh, qrsh, qlogin and qalter only.
343 Defines or redefines the path used for the standard error stream
345 of the job. For qsh, qrsh and qlogin only the standard error
346 stream of prolog and epilog is redirected. If the path consti‐
347 tutes an absolute path name, the error-path attribute of the job
348 is set to path, including the hostname. If the path name is rel‐
349 ative, Grid Engine expands path either with the current working
350 directory path (if the -cwd switch (see above) is also speci‐
351 fied) or with the home directory path. If hostname is present,
352 the standard error stream will be placed in the corresponding
353 location only if the job runs on the specified host. If the path
354 contains a ":" without a hostname, a leading ":" has to be spec‐
355 ified.
356 By default the file name for interactive jobs is /dev/null. For
358 batch jobs the default file name has the form job_name.ejob_id
359 and job_name.ejob_id.task_id for array job tasks (see -t option
360 below).
361 If path is a directory, the standard error stream of the job
363 will be put in this directory under the default file name. If
364 the pathname contains certain pseudo-environment-variables,
365 their value will be expanded at runtime of the job and will be
366 used to constitute the standard error stream path name. The fol‐
367 lowing pseudo-environment-variables are supported currently:
368 $HOME home directory on execution machine
370 $USER user ID of job owner
371 $JOB_ID current job ID
372 $JOB_NAME current job name (see -N option)
373 $HOSTNAME name of the execution host
374 $TASK_ID array job task index number
375 The tilde sign ("~") can be used as an alternative to $HOME,
377 as is common in csh(1) or ksh(1). Note, that the "~" sign also
378 works in combination with user names, so that "~<user>" expands
379 to the home directory of <user>. Using another user ID than that
380 of the job owner requires corresponding permissions, of course.
381 qalter allows changing this option even while the job executes.
383 The modified parameter will only be in effect after a restart or
384 migration of the job, however.
385 -hard Available for qsub, qsh, qrsh, qlogin and qalter only.
387 Signifies that all resource requirements following in the com‐
389 mand line will be hard requirements and must be satisfied in
390 full before a job can be scheduled.
391 As Grid Engine scans the command line and script file for Grid
392 Engine options and parameters it builds a list of resources
393 required by a job. All such resource requests are considered as
394 absolutely essential for the job to commence. If the -soft
395 option (see below) is encountered during the scan then all fol‐
396 lowing resources are designated as "soft requirements" for exe‐
397 cution, or "nice-to-have, but not essential". If the -hard flag
398 is encountered at a later stage of the scan, all resource
399 requests following it once again become "essential". The -hard
400 and -soft options in effect act as "toggles" during the scan.
401 -h | -h {u|s|o|n|U|O|S}...
403 Available for qsub, qsh, qrsh, qlogin, qalter and qresub.
404 List of holds to place on a job, a task or some tasks of a job.
406 `u' denotes a user hold.
408 `s' denotes a system hold.
409 `o' denotes a operator hold.
410 `n' denotes no hold.
411 As long as any hold other than `n' is assigned to the job the
413 job is not eligible for execution. Holds can be released via
414 qalter and qrls(1). In case of qalter this is supported by the
415 following additional option specifiers for the -h switch:
416 `U' removes a user hold.
418 `S' removes a system hold.
419 `O' removes a operator hold.
420 Grid Engine managers can assign and remove all hold types, Grid
422 Engine operators can assign and remove user and operator holds,
423 and users can only assign or remove user holds.
424 In the case of qsub only user holds can be placed on a job and
426 thus only the first form of the option with the -h switch alone
427 is allowed. As opposed to this, qalter requires the second form
428 described above.
429 An alternate means to assign hold is provided by the qhold(1)
431 facility.
432 If the job is an array job (see the -t option below), all tasks
434 specified via -t are affected by the -h operation simultane‐
435 ously.
436 qalter allows changing this option even while the job is execut‐
438 ing. The modified parameter will only be in effect after a
439 restart or migration of the job, however.
440 -help Prints a listing of all options.
442 -hold_jid wc_job_list
444 Available for qsub, qrsh, and qalter only. See sge_types(1).
445 for 504
446 Defines or redefines the job dependency list of the submitted
448 job. A reference by job name or pattern is only accepted if the
449 referenced job is owned by the same user as the referring job.
450 The submitted job is not eligible for execution unless all jobs
451 referenced in the comma-separated job id and/or job name list
452 have completed. If any of the referenced jobs exits with exit
453 code 100, the submitted job will remain ineligible for execu‐
454 tion.
455 With the help of job names or regular pattern one can specify a
457 job dependency on multiple jobs satisfying the regular pattern
458 or on all jobs with the requested name. The name dependencies
459 are resolved at submit time and can only be changed via qalter.
460 New jobs or name changes of other jobs will not be taken into
461 account.
462 qalter allows changing this option even while the job executes.
464 The modified parameter will only be in effect after a restart or
465 migration of the job, however.
466 -i [[hostname]:]file,...
468 Available for qsub, and qalter only.
469 Defines or redefines the file used for the standard input stream
471 of the job. If the file constitutes an absolute filename, the
472 input-path attribute of the job is set to path, including the
473 hostname. If the path name is relative, Grid Engine expands path
474 either with the current working directory path (if the -cwd
475 switch (see above) is also specified) or with the home directory
476 path. If hostname is present, the standard input stream will be
477 placed in the corresponding location only if the job runs on the
478 specified host. If the path contains a ":" without a hostname, a
479 leading ":" has to be specified.
480 By default /dev/null is the input stream for the job.
482 It is possible to use certain pseudo-variables, whose values
484 will be expanded at time the job runs and will be used to
485 express the standard input stream as described in the -e option
486 for the standard error stream.
487 qalter allows changing this option even while the job executes.
489 The modified parameter will only be in effect after a restart or
490 migration of the job, however.
491 -inherit
493 Available only for qrsh and qmake(1).
494 qrsh allows the user to start a task in an already scheduled
496 parallel job. The option -inherit tells qrsh to read a job id
497 from the environment variable JOB_ID and start the specified
498 command as a task in this job. Please note that in this case,
499 the hostname of the host where the command will be executed must
500 precede the command to execute; the syntax changes to
501 qrsh -inherit [ other options ] hostname command [ command_args
503 ]
504 Note also, that in combination with -inherit, most other command
506 line options will be ignored. Only the options -verbose, -v and
507 -V will be interpreted. As a replacement to option -cwd please
508 use -v PWD.
509 Usually a task should have the same environment (including the
511 current working directory) as the corresponding job, so specify‐
512 ing the option -V should be suitable for most applications.
513 Note: If in your system the qmaster tcp port is not configured
515 as a service, but rather via the environment variable SGE_QMAS‐
516 TER_PORT, make sure that this variable is set in the environment
517 when calling qrsh or qmake with the -inherit option. If you call
518 qrsh or qmake with the -inherit option from within a job script,
519 export SGE_QMASTER_PORT with the option "-v SGE_QMASTER_PORT"
520 either as a command argument or an embedded directive.
521 -j y[es]|n[o]
523 Available for qsub, qsh, qrsh, qlogin and qalter only.
524 Specifies whether or not the standard error stream of the job is
526 merged into the standard output stream.
527 If both the -j y and the -e options are present, Grid Engine
528 sets but ignores the error-path attribute.
529 qalter allows changing this option even while the job executes.
531 The modified parameter will only be in effect after a restart or
532 migration of the job, however.
533 -js job_share
535 Available for qsub, qsh, qrsh, qlogin and qalter only.
536 Defines or redefines the job share of the job relative to other
538 jobs. Job share is an unsigned integer value. The default job
539 share value for jobs is 0.
540 The job share influences the Share Tree Policy and the Func‐
542 tional Policy. It has no effect on the Urgency and Override
543 Policies (see share_tree(5), sched_conf(5) and the Grid Engine
544 Installation and Administration Guide for further information on
545 the resource management policies supported by Grid Engine).
546 In case of the Share Tree Policy, users can distribute the tick‐
548 ets to which they are currently entitled among their jobs using
549 different shares assigned via -js. If all jobs have the same job
550 share value, the tickets are distributed evenly. Otherwise, jobs
551 receive tickets relative to the different job shares. Job shares
552 are treated like an additional level in the share tree in the
553 latter case.
554 In connection with the Functional Policy, the job share can be
556 used to weight jobs within the functional job category. Tickets
557 are distributed relative to any uneven job share distribution
558 treated as a virtual share distribution level underneath the
559 functional job category.
560 If both the Share Tree and the Functional Policy are active, the
562 job shares will have an effect in both policies, and the tickets
563 independently derived in each of them are added to the total
564 number of tickets for each job.
565 -l resource=value,...
567 Available for qsub, qsh, qrsh, qlogin and qalter only.
568 Launch the job in a Grid Engine queue meeting the given resource
570 request list. In case of qalter the previous definition is
571 replaced by the specified one.
572 complex(5) describes how a list of available resources and their
573 associated valid value specifiers can be obtained.
574 There may be multiple -l switches in a single command. You may
575 request multiple -l options to be soft or hard both in the same
576 command line. In case of a serial job multiple -l switches
577 refine the definition for the sought queue.
578 qalter allows changing this option even while the job executes.
580 The modified parameter will only be in effect after a restart or
581 migration of the job, however.
582 -m b|e|a|s|n,...
584 Available for qsub, qsh, qrsh, qlogin and qalter only.
585 Defines or redefines under which circumstances mail is to be
587 sent to the job owner or to the users defined with the -M option
588 described below. The option arguments have the following mean‐
589 ing:
590 `b' Mail is sent at the beginning of the job.
592 `e' Mail is sent at the end of the job.
593 `a' Mail is sent when the job is aborted or
594 rescheduled.
595 `s' Mail is sent when the job is suspended.
596 `n' No mail is sent.
597 By default, no mail is sent when a job is suspended.
599 qalter allows changing the b, e, and a option arguments even
601 while the job executes. The modification of the b option argu‐
602 ment will only be in effect after a restart or migration of the
603 job, however.
604 -M user[@host],...
606 Available for qsub, qsh, qrsh, qlogin and qalter only.
607 Defines or redefines the list of users to which the server that
609 executes the job has to send mail, if the server sends mail
610 about the job.
611 qalter allows changing this option even while the job executes.
613 -masterq wc_queue_list
615 Available for qsub, qrsh, qsh, qlogin and qalter. Only meaning‐
616 ful for parallel jobs, i.e. together with the -pe option.
617 Defines or redefines a list of cluster queues, queue domains and
619 queue instances which may be used to become the master queue of
620 this parallel job. A more detailed description of wc_queue_list
621 can be found in sge_types(1). The master queue is defined as
622 the queue where the parallel job is started. The other queues to
623 which the parallel job spawns tasks are called slave queues. A
624 parallel job only has one master queue.
625 This parameter has all the properties of a resource request and
627 will be merged with requirements derived from the -l option
628 described above.
629 qalter allows changing this option even while the job executes.
631 The modified parameter will only be in effect after a restart or
632 migration of the job, however.
633 -notify
635 Available for qsub, qrsh (with command) and qalter only.
636 This flag, when set causes Grid Engine to send "warning" signals
638 to a running job prior to sending the signals themselves. If a
639 SIGSTOP is pending, the job will receive a SIGUSR1 several sec‐
640 onds before the SIGSTOP. If a SIGKILL is pending, the job will
641 receive a SIGUSR2 several seconds before the SIGKILL. The
642 amount of time delay is controlled by the notify parameter in
643 each queue configuration (see queue_conf(5)).
644 Note that the Linux operating system "misused" the user signals
646 SIGUSR1 and SIGUSR2 in some early Posix thread implementations.
647 You might not want to use the -notify option if you are running
648 multi-threaded applications in your jobs under Linux, particu‐
649 larly on 2.0 or earlier kernels.
650 qalter allows changing this option even while the job executes.
652 -now y[es]|n[o]
654 Available for qsub, qsh, qlogin and qrsh.
655 -now y tries to start the job immediately or not at all. The
657 command returns 0 on success, or 1 on failure (also if the job
658 could not be scheduled immediately). For array jobs submitted
659 with the -now option, if all tasks cannot be immediately sched‐
660 uled, no tasks are scheduled. -now y is default for qsh, qlogin
661 and qrsh
662 With the -now n option, the job will be put into the pending
663 queue if it cannot be executed immediately. -now n is default
664 for qsub.
665 -N name
667 Available for qsub, qsh, qrsh, qlogin and qalter only.
668 The name of the job. The name can be any printable set of char‐
670 acters except "\n", "\t", "\r", "/", ":", "@", "\", "*", and
671 "?", and it has to start with an alphabetic character. Invalid
672 job names will be denied at submit time.
673 If the -N option is not present, Grid Engine assigns the name of
674 the job script to the job after any directory pathname has been
675 removed from the script-name. If the script is read from stan‐
676 dard input, the job name defaults to STDIN.
677 If qsh or qlogin is run without the -N option, the string
678 `INTERACT' is assigned to the job.
679 If qrsh is run without the -N option, the resulting job name is
680 determined from the qrsh command line by using the argument
681 string up to the first occurrence of a semicolon or whitespace
682 and removing the directory pathname.
683 qalter allows changing this option even while the job executes.
685 -noshell
687 Available only for qrsh with a command line.
688 Do not start the command line given to qrsh in a user's login
690 shell, i.e. execute it without the wrapping shell.
691 This option can be used to speed up execution as some overhead,
693 like the shell startup and sourcing the shell resource files, is
694 avoided.
695 This option can only be used if no shell-specific command line
697 parsing is required. If the command line contains shell syntax
698 like environment variable substitution or (back) quoting, a
699 shell must be started. In this case, either do not use the
700 -noshell option or include the shell call in the command line.
701 Example:
703 qrsh echo '$HOSTNAME'
704 Alternative call with the -noshell option
705 qrsh -noshell /bin/tcsh -f -c 'echo $HOSTNAME'
706 -nostdin
708 Available only for qrsh.
709 Suppress the input stream STDIN - qrsh will pass the option -n
711 to the rsh(1) command. This is especially useful, if multiple
712 tasks are executed in parallel using qrsh, e.g. in a make(1)
713 process - it would be undefined, which process would get the
714 input.
715 -o [[hostname]:]path,...
717 Available for qsub, qsh, qrsh, qlogin and qalter only.
718 The path used for the standard output stream of the job. The
720 path is handled as described in the -e option for the standard
721 error stream.
722 By default the file name for standard output has the form
724 job_name.ojob_id and job_name.ojob_id.task_id for array job
725 tasks (see -t option below).
726 qalter allows changing this option even while the job executes.
728 The modified parameter will only be in effect after a restart or
729 migration of the job, however.
730 -ot override_tickets
732 Available for qalter only.
733 Changes the number of override tickets for the specified job.
735 Requires manager/operator privileges.
736 -P project_name
738 Available for qsub, qsh, qrsh, qlogin and qalter only.
739 Specifies the project to which this job is assigned. The admin‐
741 istrator needs to give permission to individual users to submit
742 jobs to a specific project. (see -aprj option to qconf(1)).
743 -p priority
745 Available for qsub, qsh, qrsh, qlogin and qalter only.
746 Defines or redefines the priority of the job relative to other
748 jobs. Priority is an integer in the range -1023 to 1024. The
749 default priority value for jobs is 0.
750 Users may only decrease the priority of their jobs. Grid Engine
752 managers and administrators may also increase the priority asso‐
753 ciated with jobs. If a pending job has higher priority, it is
754 earlier eligible for being dispatched by the Grid Engine sched‐
755 uler.
756 -pe parallel_environment n[-[m]]|[-]m,...
759 Available for qsub, qsh, qrsh, qlogin and qalter only.
760 Start parallel programming environment (PE). The range descrip‐
762 tor behind the PE name specifies the number of parallel pro‐
763 cesses to be run. Grid Engine will allocate the appropriate
764 resources as available. The sge_pe(5) manual page contains
765 information about the definition of PEs and about how to obtain
766 a list of currently valid PEs.
767 You can specify a PE name which uses the wildcard character,
768 "*". Thus the request "pvm*" will match any parallel environ‐
769 ment with a name starting with the string "pvm". In the case of
770 multiple parallel environments whose names match the name
771 string, the parallel environment with the most available slots
772 is chosen.
773 The range specification is a list of range expressions of the
774 form "n-m", where n and m are positive, non-zero integers. The
775 form "n" is equivalent to "n-n". The form "-m" is equivalent to
776 "1-m". The form "n-" is equivalent to "n-infinity". The range
777 specification is processed as follows: The largest number of
778 queues requested is checked first. If enough queues meeting the
779 specified attribute list are available, all are allocated. If
780 not, the next smaller number of queues is checked, and so forth.
781 If additional -l options are present, they restrict the set of
782 eligible queues for the parallel job.
783 qalter allows changing this option even while the job executes.
785 The modified parameter will only be in effect after a restart or
786 migration of the job, however.
787 -q wc_queue_list
789 Available for qsub, qrsh, qsh, qlogin and qalter.
790 Defines or redefines a list of cluster queues, queue domains or
792 queue instances which may be used to execute this job. Please
793 find a description of wc_queue_list in sge_types(1). This
794 parameter has all the properties of a resource request and will
795 be merged with requirements derived from the -l option described
796 above.
797 qalter allows changing this option even while the job executes.
799 The modified parameter will only be in effect after a restart or
800 migration of the job, however.
801 -R y[es]|n[o]
803 Available for qsub, qrsh, qsh, qlogin and qalter.
804 Indicates whether a reservation for this job should be done.
806 Reservation is never done for immediate jobs, i.e. jobs submit‐
807 ted using the -now yes option. Please note that regardless of
808 the reservation request, job reservation might be disabled using
809 max_reservation in sched_conf(5) and might be limited only to a
810 certain number of high priority jobs.
811 By default jobs are submitted with the -R n option.
813 -r y[es]|n[o]
815 Available for qsub and qalter only.
816 Identifies the ability of a job to be rerun or not. If the
818 value of -r is 'yes', the job will be rerun if the job was
819 aborted without leaving a consistent exit state. (This is typi‐
820 cally the case if the node on which the job is running crashes).
821 If -r is 'no', the job will not be rerun under any circum‐
822 stances.
823 Interactive jobs submitted with qsh, qrsh or qlogin are not re-
824 runnable.
825 qalter allows changing this option even while the job executes.
827 -sc variable[=value],...
831 Available for qsub, qsh, qrsh, qlogin and qalter only.
832 Sets the given name/value pairs as the job's context. Value may
834 be omitted. Grid Engine replaces the job's previously defined
835 context with the one given as the argument. Multiple -ac, -dc,
836 and -sc options may be given. The order is important.
837 Contexts provide a way to dynamically attach and remove meta-
838 information to and from a job. The context variables are not
839 passed to the job's execution context in its environment.
840 qalter allows changing this option even while the job executes.
842 -shell y[es]|n[o]
844 Available only for qsub.
845 -shell n causes qsub to execute the command line directly, as if
847 by exec(2). No command shell will be executed for the job.
848 This option only applies when -b y is also used. Without -b y,
849 -shell n has no effect.
850 This option can be used to speed up execution as some overhead,
852 like the shell startup and sourcing the shell resource files is
853 avoided.
854 This option can only be used if no shell-specific command line
856 parsing is required. If the command line contains shell syntax,
857 like environment variable substitution or (back) quoting, a
858 shell must be started. In this case either do not use the
859 -shell n option or execute the shell as the command line and
860 pass the path to the executable as a parameter.
861 If a job executed with the -shell n option fails due to a user
863 error, such as an invalid path to the executable, the job will
864 enter the error state.
865 -shell y cancels the effect of a previous -shell n. Otherwise,
867 it has no effect.
868 See -b and -noshell for more information.
870 -soft Available for qsub, qsh, qrsh, qlogin and qalter only.
872 Signifies that all resource requirements following in the com‐
874 mand line will be soft requirements and are to be filled on an
875 "as available" basis.
876 As Grid Engine scans the command line and script file for Grid
877 Engine options and parameters, it builds a list of resources
878 required by the job. All such resource requests are considered
879 as absolutely essential for the job to commence. If the -soft
880 option is encountered during the scan then all following
881 resources are designated as "soft requirements" for execution,
882 or "nice-to-have, but not essential". If the -hard flag (see
883 above) is encountered at a later stage of the scan, all resource
884 requests following it once again become "essential". The -hard
885 and -soft options in effect act as "toggles" during the scan.
886 -sync y[es]|n[o]
888 Available for qsub.
889 -sync y causes qsub to wait for the job to complete before exit‐
891 ing. If the job completes successfully, qsub's exit code will
892 be that of the completed job. If the job fails to complete suc‐
893 cessfully, qsub will print out a error message indicating why
894 the job failed and will have an exit code of 1. If qsub is
895 interrupted, e.g. with CTRL-C, before the job completes, the job
896 will be canceled.
897 With the -sync n option, qsub will exit with an exit code of 0
898 as soon as the job is submitted successfully. -sync n is
899 default for qsub.
900 If -sync y is used in conjunction with -now y, qsub will behave
901 as though only -now y were given until the job has been success‐
902 fully scheduled, after which time qsub will behave as though
903 only -sync y were given.
904 If -sync y is used in conjunction with -t n[-m[:i]], qsub will
905 wait for all the job's tasks to complete before exiting. If all
906 the job's tasks complete successfully, qsub's exit code will be
907 that of the first completed job tasks with a non-zero exit code,
908 or 0 if all job tasks exited with an exit code of 0. If any of
909 the job's tasks fail to complete successfully, qsub will print
910 out an error message indicating why the job task(s) failed and
911 will have an exit code of 1. If qsub is interrupted, e.g. with
912 CTRL-C, before the job completes, all of the job's tasks will be
913 canceled.
914 -S [[hostname]:]pathname,...
916 Available for qsub, qsh and qalter.
917 Specifies the interpreting shell for the job. Only one pathname
919 component without a host specifier is valid and only one path
920 name for a given host is allowed. Shell paths with host assign‐
921 ments define the interpreting shell for the job if the host is
922 the execution host. The shell path without host specification is
923 used if the execution host matches none of the hosts in the
924 list.
925 Furthermore, the pathname can be constructed with pseudo-envi‐
927 ronment-variables as described for the -e option above.
928 In the case of qsh the specified shell path is used to execute
930 the corresponding command interpreter in the xterm(1) (via its
931 -e option) started on behalf of the interactive job.
932 qalter allows changing this option even while the job executes.
934 The modified parameter will only be in effect after a restart or
935 migration of the job, however.
936 -t n[-m[:s]]
938 Available for qsub and qalter only.
939 Submits a so called Array Job, i.e. an array of identical tasks
941 being differentiated only by an index number and being treated
942 by Grid Engine almost like a series of jobs. The option argument
943 to -t specifies the number of array job tasks and the index num‐
944 ber which will be associated with the tasks. The index numbers
945 will be exported to the job tasks via the environment variable
946 SGE_TASK_ID. The option arguments n, m and s will be available
947 through the environment variables SGE_TASK_FIRST, SGE_TASK_LAST
948 and SGE_TASK_STEPSIZE.
949 The following restrictions apply to the values n and m:
951 1 <= n <= MIN(2^31-1, max_aj_tasks)
953 1 <= m <= MIN(2^31-1, max_aj_tasks)
954 n <= m
955 max_aj_tasks is defined in the cluster configuration (see
957 sge_conf(5))
958 The task id range specified in the option argument may be a sin‐
960 gle number, a simple range of the form n-m or a range with a
961 step size. Hence, the task id range specified by 2-10:2 would
962 result in the task id indexes 2, 4, 6, 8, and 10, for a total of
963 5 identical tasks, each with the environment variable
964 SGE_TASK_ID containing one of the 5 index numbers.
965 All array job tasks inherit the same resource requests and
967 attribute definitions as specified in the qsub or qalter command
968 line, except for the -t option. The tasks are scheduled indepen‐
969 dently and, provided enough resources exist, concurrently, very
970 much like separate jobs. An array job or a sub-array can be
971 accessed, however, as a single unit by commands like qmod(1) or
972 qdel(1). See the corresponding manual pages for further detail.
973 Array jobs are commonly used to execute the same type of opera‐
975 tion on varying input data sets correlated with the task index
976 number. The number of tasks in a array job is unlimited.
977 STDOUT and STDERR of array job tasks will be written into dif‐
979 ferent files with the default location
980 <jobname>.['e'|'o']<job_id>'.'<task_id>
982 In order to change this default, the -e and -o options (see
984 above) can be used together with the pseudo-environment-vari‐
985 ables $HOME, $USER, $JOB_ID, $JOB_NAME, $HOSTNAME, and
986 $SGE_TASK_ID.
987 Note, that you can use the output redirection to divert the out‐
989 put of all tasks into the same file, but the result of this is
990 undefined.
991 -terse
995 Available for qsub only.
996 Causes qsub to display only the job-id of the job being submit‐
998 ted rather than the regular "Your job ..." string. In case of
999 an error the error is reported on stderr as usual.
1000 This can be helpful for scripts which need to parse qsub output
1001 to get the job-id.
1002 -u username,...
1004 Available for qalter only. Changes are only made on those jobs
1005 which were submitted by users specified in the list of user‐
1006 names. For managers it is possible to use the qalter -u '*'
1007 command to modify all jobs of all users.
1008 If you use the -u switch it is not permitted to specify an addi‐
1010 tional wc_job_range_list.
1011 -v variable[=value],...
1013 Available for qsub, qsh, qrsh and qalter.
1014 Defines or redefines the environment variables to be exported to
1016 the execution context of the job. If the -v option is present
1017 Grid Engine will add the environment variables defined as argu‐
1018 ments to the switch and, optionally, values of specified vari‐
1019 ables, to the execution context of the job.
1020 qalter allows changing this option even while the job executes.
1022 The modified parameter will only be in effect after a restart or
1023 migration of the job, however.
1024 -verbose
1026 Available only for qrsh and qmake(1).
1027 Unlike qsh and qlogin, qrsh does not output any informational
1029 messages while establishing the session, compliant with the
1030 standard rsh(1) and rlogin(1) system calls. If the option -ver‐
1031 bose is set, qrsh behaves like the qsh and qlogin commands,
1032 printing information about the process of establishing the
1033 rsh(1) or rlogin(1) session.
1034 -verify
1036 Available for qsub, qsh, qrsh, qlogin and qalter.
1037 Instead of submitting a job, this option prints detailed infor‐
1039 mation about the would-be job as though qstat(1) -j were used,
1040 including the effects of command-line parameters and the exter‐
1041 nal environment.
1042 -V Available for qsub, qsh, qrsh with command, qalter and qresub.
1044 Specifies that all environment variables active within the qsub
1046 utility be exported to the context of the job.
1047 -w e|w|n|v
1049 Available for qsub, qsh, qrsh, qlogin and qalter.
1050 Specifies a validation level applied to the job to be submitted
1052 (qsub, qlogin, and qsh) or the specified queued job (qalter).
1053 The information displayed indicates whether the job can possibly
1054 be scheduled assuming an empty system with no other jobs.
1055 Resource requests exceeding the configured maximal thresholds or
1056 requesting unavailable resource attributes are possible causes
1057 for jobs to fail this validation.
1058 The specifiers e, w, n and v define the following validation
1060 modes:
1061 `e' error - jobs with invalid requests will be
1063 rejected; the default for qrsh, qsh
1064 and qlogin.
1065 `w' warning - only a warning will be displayed
1066 for invalid requests.
1067 `n' none - switches off validation; the default for
1068 qsub and qalter.
1069 `v' verify - does not submit the job but prints an
1070 extensive validation report.
1071 Note, that the necessary checks are performance consuming and
1073 hence the checking is switched off by default.
1074 Note also, that the reasons for job requirements being invalid
1076 with respect to resource availability of queues are displayed in
1077 the "-w v" case using the format as described for the qstat(1)
1078 -F option (see description of Full Format in section OUTPUT FOR‐
1079 MATS of the qstat(1) manual page.
1080 -wd working_dir
1082 Available for qsub, qsh, qrsh and qalter only.
1083 Execute the job from the directory specified in working_dir.
1085 This switch will activate Grid Engine's path aliasing facility,
1086 if the corresponding configuration files are present (see
1087 sge_aliases(5)).
1088 qalter allows changing this option even while the job executes.
1090 The modified parameter will only be in effect after a restart or
1091 migration of the job, however.
1092 command
1094 Available for qsub and qrsh only.
1095 The job's scriptfile or binary. If not present or if the oper‐
1097 and is the single-character string '-', qsub reads the script
1098 from standard input.
1099 command_args
1101 Available for qsub, qrsh and qalter only.
1102 Arguments to the job. Not valid if the script is entered from
1104 standard input.
1105 qalter allows changing this option even while the job executes.
1107 The modified parameter will only be in effect after a restart or
1108 migration of the job, however.
1109 xterm_args
1111 Available for qsh only.
1112 Arguments to the xterm(1) executable, as defined in the configu‐
1114 ration. For details, refer to sge_conf(5)).
1115
1117 SGE_ROOT Specifies the location of the Grid Engine standard con‐
1118 figuration files.
1119 SGE_CELL If set, specifies the default Grid Engine cell. To
1121 address a Grid Engine cell qsub, qsh, qlogin or qalter
1122 use (in the order of precedence):
1123 The name of the cell specified in the environment
1125 variable SGE_CELL, if it is set.
1126 The name of the default cell, i.e. default.
1128 SGE_DEBUG_LEVEL
1131 If set, specifies that debug information should be writ‐
1132 ten to stderr. In addition the level of detail in which
1133 debug information is generated is defined.
1134 SGE_QMASTER_PORT
1136 If set, specifies the tcp port on which sge_qmaster(8)
1137 is expected to listen for communication requests. Most
1138 installations will use a services map entry for the ser‐
1139 vice "sge_qmaster" instead to define that port.
1140 DISPLAY For qsh jobs the DISPLAY has to be specified at job sub‐
1142 mission. If the DISPLAY is not set by using the -dis‐
1143 play or the -v switch, the contents of the DISPLAY envi‐
1144 ronment variable are used as default.
1145 In addition to those environment variables specified to be exported to
1147 the job via the -v or the -V option (see above) qsub, qsh, and qlogin
1148 add the following variables with the indicated values to the variable
1149 list:
1150 SGE_O_HOME the home directory of the submitting client.
1153 SGE_O_HOST the name of the host on which the submitting client is
1155 running.
1156 SGE_O_LOGNAME the LOGNAME of the submitting client.
1158 SGE_O_MAIL the MAIL of the submitting client. This is the mail
1160 directory of the submitting client.
1161 SGE_O_PATH the executable search path of the submitting client.
1163 SGE_O_SHELL the SHELL of the submitting client.
1165 SGE_O_TZ the time zone of the submitting client.
1167 SGE_O_WORKDIR the absolute path of the current working directory of
1169 the submitting client.
1170 Furthermore, Grid Engine sets additional variables into the job's envi‐
1172 ronment, as listed below.
1173 SGE_ARCH The Grid Engine architecture name of the node on which
1175 the job is running. The name is compiled into the
1176 sge_execd(8) binary.
1177 SGE_CKPT_ENV Specifies the checkpointing environment (as selected
1179 with the -ckpt option) under which a checkpointing job
1180 executes. Only set for checkpointing jobs.
1181 SGE_CKPT_DIR Only set for checkpointing jobs. Contains path ckpt_dir
1183 (see checkpoint(5) ) of the checkpoint interface.
1184 SGE_STDERR_PATH
1186 the pathname of the file to which the standard error
1187 stream of the job is diverted. Commonly used for enhanc‐
1188 ing the output with error messages from prolog, epilog,
1189 parallel environment start/stop or checkpointing
1190 scripts.
1191 SGE_STDOUT_PATH
1193 the pathname of the file to which the standard output
1194 stream of the job is diverted. Commonly used for enhanc‐
1195 ing the output with messages from prolog, epilog, paral‐
1196 lel environment start/stop or checkpointing scripts.
1197 SGE_STDIN_PATH the pathname of the file from which the standard input
1199 stream of the job is taken. This variable might be used
1200 in combination with SGE_O_HOST in prolog/epilog scripts
1201 to transfer the input file from the submit to the execu‐
1202 tion host.
1203 SGE_JOB_SPOOL_DIR
1205 The directory used by sge_shepherd(8) to store job
1206 related data during job execution. This directory is
1207 owned by root or by a Grid Engine administrative account
1208 and commonly is not open for read or write access to
1209 regular users.
1210 SGE_TASK_ID The index number of the current array job task (see -t
1212 option above). This is an unique number in each array
1213 job and can be used to reference different input data
1214 records, for example. This environment variable is set
1215 to "undefined" for non-array jobs. It is possible to
1216 change the predefined value of this variable with -v or
1217 -V (see options above).
1218 SGE_TASK_FIRST The index number of the first array job task (see -t
1220 option above). It is possible to change the predefined
1221 value of this variable with -v or -V (see options
1222 above).
1223 SGE_TASK_LAST The index number of the last array job task (see -t
1225 option above). It is possible to change the predefined
1226 value of this variable with -v or -V (see options
1227 above).
1228 SGE_TASK_STEPSIZE
1230 The stepsize of the array job specification (see -t
1231 option above). It is possible to change the predefined
1232 value of this variable with -v or -V (see options
1233 above).
1234 ENVIRONMENT The ENVIRONMENT variable is set to BATCH to identify
1236 that the job is being executed under Grid Engine con‐
1237 trol.
1238 HOME The user's home directory path from the passwd(5) file.
1240 HOSTNAME The hostname of the node on which the job is running.
1242 JOB_ID A unique identifier assigned by the sge_qmaster(8) when
1244 the job was submitted. The job ID is a decimal integer
1245 in the range 1 to 99999.
1246 JOB_NAME The job name. For batch jobs or jobs submitted by qrsh
1248 with a command, the job name is built as basename of the
1249 qsub script filename resp. the qrsh command. For inter‐
1250 active jobs it is set to `INTERACTIVE' for qsh jobs,
1251 `QLOGIN' for qlogin jobs and `QRLOGIN' for qrsh jobs
1252 without a command.
1253 This default may be overwritten by the -N. option.
1255 LOGNAME The user's login name from the passwd(5) file.
1257 NHOSTS The number of hosts in use by a parallel job.
1259 NQUEUES The number of queues allocated for the job (always 1 for
1261 serial jobs).
1262 NSLOTS The number of queue slots in use by a parallel job.
1264 PATH A default shell search path of:
1266 /usr/local/bin:/usr/ucb:/bin:/usr/bin
1267 SGE_BINARY_PATH
1269 The path where the Grid Engine binaries are installed.
1270 The value is the concatenation of the cluster configura‐
1271 tion value binary_path and the architecture name
1272 $SGE_ARCH environment variable.
1273 PE The parallel environment under which the job executes
1275 (for parallel jobs only).
1276 PE_HOSTFILE The path of a file containing the definition of the vir‐
1278 tual parallel machine assigned to a parallel job by Grid
1279 Engine. See the description of the $pe_hostfile parame‐
1280 ter in sge_pe(5) for details on the format of this file.
1281 The environment variable is only available for parallel
1282 jobs.
1283 QUEUE The name of the cluster queue in which the job is run‐
1285 ning.
1286 REQUEST Available for batch jobs only.
1288 The request name of a job as specified with the -N
1290 switch (see above) or taken as the name of the job
1291 script file.
1292 RESTARTED This variable is set to 1 if a job was restarted either
1294 after a system crash or after a migration in case of a
1295 checkpointing job. The variable has the value 0 other‐
1296 wise.
1297 SHELL The user's login shell from the passwd(5) file. Note:
1299 This is not necessarily the shell in use for the job.
1300 TMPDIR The absolute path to the job's temporary working direc‐
1302 tory.
1303 TMP The same as TMPDIR; provided for compatibility with NQS.
1305 TZ The time zone variable imported from sge_execd(8) if
1307 set.
1308 USER The user's login name from the passwd(5) file.
1310
1312 There is no controlling terminal for batch jobs under Grid Engine, and
1313 any tests or actions on a controlling terminal will fail. If these
1314 operations are in your .login or .cshrc file, they may cause your job
1315 to abort.
1316 Insert the following test before any commands that are not pertinent to
1318 batch jobs in your .login:
1319 if ( $?JOB_NAME) then
1321 echo "Grid Engine spooled job"
1322 exit 0
1323 endif
1324 Don't forget to set your shell's search path in your shell start-up
1326 before this code.
1327
1329 The following exit values are returned:
1330 0 Operation was executed successfully.
1332 25 It was not possible to register a new job according to the config‐
1334 ured max_u_jobs or max_jobs limit. Additional information may be
1335 found in sge_conf(5)
1336 >0 Error occurred.
1338
1340 The following is the simplest form of a Grid Engine script file.
1341 =====================================================
1343 #!/bin/csh
1346 a.out
1347 =====================================================
1350 The next example is a more complex Grid Engine script.
1352 =====================================================
1354 #!/bin/csh
1356 # Which account to be charged cpu time
1358 #$ -A santa_claus
1359 # date-time to run, format [[CC]yy]MMDDhhmm[.SS]
1361 #$ -a 12241200
1362 # to run I want 6 or more parallel processes
1364 # under the PE pvm. the processes require
1365 # 128M of memory
1366 #$ -pe pvm 6- -l mem=128
1367 # If I run on dec_x put stderr in /tmp/foo, if I
1369 # run on sun_y, put stderr in /usr/me/foo
1370 #$ -e dec_x:/tmp/foo,sun_y:/usr/me/foo
1371 # Send mail to these users
1373 #$ -M santa@nothpole,claus@northpole
1374 # Mail at beginning/end/on suspension
1376 #$ -m bes
1377 # Export these environmental variables
1379 #$ -v PVM_ROOT,FOOBAR=BAR
1380 # The job is located in the current
1382 # working directory.
1383 #$ -cwd
1384 a.out
1386 ==========================================================
1388
1391 $REQUEST.oJID[.TASKID] STDOUT of job #JID
1392 $REQUEST.eJID[.TASKID] STDERR of job
1393 $REQUEST.poJID[.TASKID] STDOUT of par. env. of job
1394 $REQUEST.peJID[.TASKID] STDERR of par. env. of job
1395 $REQUEST.hostsJID[.TASKID] hosts file of par. env. of job
1396 $cwd/.sge_aliases cwd path aliases
1398 $cwd/.sge_request cwd default request
1399 $HOME/.sge_aliases user path aliases
1400 $HOME/.sge_request user default request
1401 $SGE_ROOT/$SGE_CELL/common/sge_aliases
1402 cluster path aliases
1403 $SGE_ROOT/$SGE_CELL/common/sge_request
1404 cluster default request
1405 $SGE_ROOT/$SGE_CELL/common/act_qmaster
1406 Grid Engine master host file
1407
1409 sge_intro(1), qconf(1), qdel(1), qhold(1), qmod(1), qrls(1), qstat(1),
1410 accounting(5), sge_aliases(5), sge_conf(5), sge_request(5), sge_pe(5),
1411 complex(5).
1412
1414 If configured correspondingly, qrsh and qlogin contain portions of the
1415 rsh, rshd, telnet and telnetd code copyrighted by The Regents of the
1416 University of California. Therefore, the following note applies with
1417 respect to qrsh and qlogin: This product includes software developed by
1418 the University of California, Berkeley and its contributors.
1419 See sge_intro(1) as well as the information provided in
1421 $SGE_ROOT/3rd_party/qrsh and $SGE_ROOT/3rd_party/qlogin for a statement
1422 of further rights and permissions.
1423GE 6.1 $Date: 2007/07/19 08:17:16 $ SUBMIT(1)