TORQUE Resource Manager
start a pbs batch execution mini-server
pbs_mom [-C chkdirectory] [-c config] [-d directory] [-h hostname]
[-L logfile] [-M MOMport] [-R RPPport] [-p|-r] [-x]
The pbs_mom command starts the operation of a batch Machine Oriented Mini-server (MOM) on the local host. Typically, this command will be in a local boot file such as /etc/rc.local. To insure that the pbs_mom command is not runnable by the general user community, the server will only execute if its real and effective uid is zero.
The first function of pbs_mom is to place jobs into execution as directed by the server, establish resource usage limits, monitor the job's usage, and notify the server when the job completes. If they exist, pbs_mom will execute a prologue script before executing a job and an epilogue script after executing the job.
The second function of pbs_mom is to respond to resource monitor requests. This was done by a separate process in previous versions of PBS but has now been combined into one process. The resource monitor function is provided mainly for the PBS scheduler. It provides information about the status of running jobs, memory available etc.
The last function of pbs_mom is to respond to task manager requests. This involves communicating with running tasks over a tcp socket as well as communicating with other MOMs within a job (a.k.a. a "sisterhood").
pbs_mom will record a diagnostic message in a log file for any error occurrence. The log files are maintained in the mom_logs directory below the home directory of the server. If the log file cannot be opened, the diagnostic message is written to the system console.
||Specifies The path of the directory used to hold checkpoint files. [Currently this is only valid on Cray systems.] The default directory is PBS_HOME/spool/checkpoint, see the -d option. The directory specified with the -C option must be owned by root and accessible (rwx) only by root to protect the security of the checkpoint files.
||Specify a alternative configuration file, see description below. If this is a relative file name it will be relative to PBS_HOME/mom_priv, see the -d option. If the specified file cannot be opened, pbs_mom will abort. If the -c option is not supplied, pbs_mom will attempt to open the default configuration file "config" in PBS_HOME/mom_priv. If this file is not present, pbs_mom will log the fact and continue.
||Set MOMs hostname. This can be useful on multi-homed networks.
||Specifies the path of the directory which is the home of the servers working files, PBS_HOME. This option is typically used along with -M when debugging MOM. The default directory is given by $PBS_SERVER_HOME which is typically /usr/spool/PBS.
||Specify an absolute path name for use as the log file. If not specified, MOM will open a file named for the current date in the PBS_HOME/mom_logs directory, see the -d option.
||Specifies the port number on which the mini-server (MOM) will listen for batch requests.
||Specifies the port number on which the mini-server (MOM) will listen for resource monitor requests, task manager requests and inter-MOM messages. Both a UDP and a TCP port of this number will be used.
||Specifies the impact on jobs which were in execution when the mini-server shut down. On any restart of MOM, the new mini-server will not be the parent of any running jobs, MOM has lost control of her offspring (not a new situation for a mother). With the -p option, MOM will allow the jobs to continue to run and monitor them indirectly via polling. This flag is redundant in that this is the default behavior when starting the server. The -p option is mutually exclusive with the -r and -q options.
||Specifies the impact on jobs which were in execution when the mini-server shut down. With the -q option, MOM will allow the processes belonging to jobs to continue to run, but will not attempt to monitor them. The -q option is mutually exclusive with the -p and -r options.
Specifies the impact on jobs which were in execution when the mini-server shut down. With the -r option, MOM will kill any processes belonging to jobs, mark the jobs as terminated, and notify the batch server which owns the job. The -r option is mutually exclusive with the -p and -q options.
Normally the mini-server is started from the system boot file without the -p or the -r option. The mini-server will make no attempt to signal the former session of any job which may have been running when the mini-server terminated. It is assumed that on reboot, all processes have been killed.
If the -r option is used following a reboot, process IDs (pids) may be reused and MOM may kill a process that is not a batch session.
||Used to specify the alarm timeout in seconds for computing a resource. Every time a resource request is processed, an alarm is set for the given amount of time. If the request has not completed before the given time, an alarm signal is generated. The default is 5 seconds.
||Disables the check for privileged port resource monitor connections. This is used mainly for testing since the privileged port is the only mechanism used to prevent any ordinary user from connecting.
The configuration file may be specified on the command line at program start with the -c flag. The use of this file is to provide several types of run time information to pbs_mom: static resource names and values, external resources provided by a program to be run on request via a shell escape, and values to pass to internal set up functions at initialization (and re-initialization).
Each item type is on a single line with the component parts separated by white space. If the line starts with a hash mark (pound sign, #), the line is considered to be a comment and is skipped.
For static resource names and values, the configuration file contains a list of resource names/values pairs, one pair per line and separated by white space. An Example of static resource names and values could be the number of tape drives of different types and could be specified by:
- tape3480 4
- tape3420 2
- tapedat 1
- tape8mm 1
If the first character of the value is an exclamation mark (!), the entire rest of the line is saved to be executed through the services of the system(3) standard library routine.
The shell escape provides a means for the resource monitor to yield arbitrary information to the scheduler. Parameter substitution is done such that the value of any qualifier sent with the query, as explained below, replaces a token with a percent sign (%) followed by the name of the qualifier. For example, here is a configuration file line which gives a resource name of "escape":
If a query for "escape" is sent with no qualifiers, the command executed would be echo %xxx %yyy. If one qualifier is sent, escape[xxx=hi there], the command executed would be echo hi there %yyy. If two qualifiers are sent, escape[xxx=hi][yyy=there], the command executed would be echo hi there. If a qualifier is sent with no matching token in the command line, escape[zzz=snafu], an error is reported.
Specifies that the available and configured disk space in the <FS> filesystem is to be reported to the pbs_server and scheduler. To request disk space on a per job basis, specify the file resource as in, qsub -l nodes=1,file=1000kb. For example, the available and configured disk space in the /localscratch filesystem will be reported:
An initialization value directive has a name which starts with a dollar sign ($) and must be known to MOM via an internal table. The entries in this table now are:
Defines hostnames running pbs_server that will be allowed to submit jobs, issue Resource Monitor (RM) requests, and get status updates. MOM will continually attempt to contact all server hosts for node status and state updates. Like $PBS_SERVER_HOME/server_name, the hostname may be followed by a colon and a port number. This parameter replaces the oft-confused $clienthost parameter from TORQUE 2.0.0p0 and earlier. Note that the hostname in $PBS_SERVER_HOME/server_name is used if no $pbsserver parameters are found.
Causes a host name to be added to the list of hosts which will be allowed to connect to MOM as long as they are using a privilaged port for the purposes of resource monitor requests. For example, here are two configuration file lines which will allow the hosts "fred" and "wilma" to connect:
Two host name are always allowed to connection to pbs_mom, "localhost" and the name returned to pbs_mom by the system call gethostname(). These names need not be specified in the configuration file. The hosts listed as "clients" can issue Resource Monitor (RM) requests. Other MOM nodes and servers do not need to be listed as clients.
Causes a host name to be added to the list of hosts which will be allowed to connect to MOM without needing to use a privilaged port. These names allow for wildcard matching. For example, here is a configuration file line which will allow queries from any host from the domain "ibm.com".
The restriction which applies to these connections is that only internal queries may be made. No resources from a config file will be found. This is to prevent any shell commands from being run by a non-root process. This parameter is generally not required except for some versions of OSX.
Specifies which directories should be staged with cp instead of rcp/scp. If a shared filesystem is available on all hosts in a cluster, this directive is used to make these filesystems known to MOM. For example, if /home is NFS mounted on all nodes in a cluster:
Sets a factor used to adjust wall time usage by to job to a common reference system. The factor is used for walltime calculations and limits the same as cputmult is used for cpu time.
Specifies the version of the config file data, a string.
Specifies the MOM interval in seconds. MOM checks each job for updated resource usages, exited processes, over-limit conditions, etc. once per interval. This value should be equal or lower to pbs_servers job_stat_rate. High values result in stale information reported to pbs_server. Low values result in increased system usage by MOM. Default is 45 seconds.
Causes MOM to report itself as state "down" to pbs_server in the event of a failed health check. This feature is EXPERIMENTAL and likely to be removed in the future. See HEALTH CHECK below.
Ideal processor load. Represents a low water mark for the load average. Nodes that are currently busy will consider itself free after falling below ideal_load.
If jobs are running, sets idea_load based on a simple expression. The expressions start with the variable t (total assigned CPUs) or c (existing CPUs), an operator (+ - / *), and followed by a float constant.
Specifies the verbosity of logging with higher numbers specifying more verbose logging. Values may range between 0 and 7.
If this is set to a value > 0 then pbs_mom will roll the current log file to log-file-name.1 when its size is greater than or equal to the value of log_file_max_size. This value is interpreted as kilo bytes.
If this is set to a value >=1 and log_file_max_size is set then pbs_mom will continue rolling the log files to log-file-name.log_file_roll_depth.
Maximum processor load. Nodes over this load average are considered busy (see ideal_load above).
If jobs are running, sets max_load based on a simple expression. The expressions start with the variable t (total assigned CPUs) or c (existing CPUs), an operator (+ - / *), and followed by a float constant.
Enable automatic restarts of MOM. If enabled, MOM will check if its binary has been updated and restart itself at a safe point when no jobs are running; thus making upgrades easier. The check is made by comparing the mtime of the pbs_mom executable. Command-line args, the process name, and the PATH env variable are preserved across restarts. It is recommended that this not be enabled in the config file, but enabled when desired with momctl (see RESOURCES for more information.)
Specifies the fully qualified pathname of the health check script to run (see HEALTH CHECK for more information).
Specifies when to run the MOM health check. The check can be either periodic, event-driver, or both. The value starts with an integer specifying the number of MOM intervals between subsequent executions of the specified health check. After the integer is an optional comma-separated list of event names. Currently supported are "jobstart" and "jobend". This value defaults to 1 with no events indicating the check is run every mom interval. (see HEALTH CHECK for more information)
Specifies maximum duration (in seconds) which the mom will wait for the job prolog or job job epilog to complete. This parameter defaults to 300 seconds (5 minutes).
Enables the ability to remotely reconfigure pbs_mom with a new config file. Default is disabled. This parameter accepts various forms of true, yes, and 1.
Specifies the number of seconds before TCP messages will time out. TCP messages include job obituaries, and TM requests if RPP is disabled. Default is 60 seconds.
Sets the directory basename for a per-job temporary directory. Before job launch, MOM will append the jobid to the tmpdir basename and create the directory. After the job exit, MOM will recursively delete it. The env variable TMPDIR will be set for all pro/epilog scripts, the job script, and TM tasks.
Directory creation and removal is done as the job owner and group, so the owner must have write permission to create the directory. If the directory already exists and is owned by the job owner, it will not be deleted after the job. If the directory already exists and is NOT owned by the job owner, the job start will be rejected.
Specifies (in seconds) how often MOM updates its status information to pbs_server. This value should correlate with the server's scheduling interval and its "node_check_rate" attribute. High values for "status_update_time" cause pbs_server to report stale information, while low values increase the load of pbs_server and the network. Default is 45 seconds.
Specifies the path to the xauth binary to enable X11 fowarding.
If set to true, then pbs_mom will ignore vmem/pvmem limit enforcement.
If set to true, then pbs_mom will ignore walltime limit enforcement.
Sets the local hostname as used by pbs_mom.
Resource Monitor queries can be made with momctls -q option to retrieve and set pbs_mom options. Any configured static resource may be retrieved with a request of the same name. These are resource requests not otherwise documented in the PBS ERS.
Forces an immediate MOM cycle.
Retrieve or set the $status_update_time parameter.
Retrieve or set the $check_poll_time parameter.
Retrieve the config version.
Retrieve or set the $jobstartblocktime parameter.
Retrieve or set the $enablemomrestart parameter.
Retrieve or set the $loglevel parameter.
Retrieve or set the EXPERIMENTAL $down_on_error parameter.
- diag0 - diag4
Retrieves various diagnostic information.
Retrieve or set the $rcpcmd parameter.
Retrieves the pbs_mom version.
The health check script is executed directly by the pbs_mom daemon under the root user id. It must be accessible from the compute node and may be a script or compiled executable program. It may make any needed system calls and execute any combination of system utilities but should not execute resource manager client commands. Also, as of TORQUE 1.0.1, the pbs_mom daemon blocks until the health check is completed and does not possess a built-in timeout. Consequently, it is advisable to keep the launch script execution time short and verify that the script will not block even under failure conditions.
If the script detects a failure, it should return the keyword Error to stdout followed by an error message. The message (up to 256 characters) immediately following the Error string will be assigned to the node attribute message of the associated node.
If the script detects a failure when run from "jobstart", then the job will be rejected. This should probably only be used with advanced schedulers like Moab so that the job can be routed to another node.
TORQUE currently ignores Error messages by default, but advanced schedulers like moab can be configured to react appropriately.
If the experimental $down_on_error MOM setting is enabled, MOM will set itself to state down and report to pbs_server; and pbs_server will report the node as "down". Additionally, the experimental "down_on_error" server attribute can be enabled which has the same effect but moves the decision to pbs_server. It is redundant to have MOMs $down_on_error and pbs_servers down_on_error features enabled. See "down_on_error" in pbs_server_attributes(7B).
Contains the hostname running pbs_server.
The default directory for configuration files, typically (/usr/spool/pbs)/mom_priv.
Directory for log files recorded by the server.
The administrative script to be run before job execution.
The administrative script to be run after job execution.
pbs_mom handles the following signals:
Causes pbs_mom to re-read its configuration file, close and reopen the log file, and reinitialize resource structures.
Results in a log file entry. The signal is used to limit the time taken by certain children processes, such as the prologue and epilogue.
- SIGINT and SIGTERM
Results in pbs_mom exiting without terminating any running jobs. This is the action for the following signals as well: SIGXCPU, SIGXFSZ, SIGCPULIM, and SIGSHUTDN.
- SIGUSR1, SIGUSR2
Causes mom to increase and decrease logging levels, respectively.
- SIGBUS, SIGFPE, SIGILL, SIGTRAP, and SIGSYS
Cause a core dump if the PBSCOREDUMP environmental variable is defined.
All other signals have their default behavior installed.
If the mini-server command fails to begin operation, the server exits with a value greater than zero.
pbs_server(8B), pbs_scheduler_basl(8B), pbs_scheduler_tcl(8B), the PBS External Reference Specification, and the PBS Administrators Guide.