helios.pl - Launch a daemon to service jobs in the Helios job processing system
export HELIOS_INI=/path/to/helios.ini [export HELIOS_DEBUG=1] helios.pl <jobclass> [--clear-halt] # just prints version info helios.pl --version
The helios.pl program, given a subclass of Helios::Service, will launch a daemon to service Helios jobs of that subclass. The number of worker processes to run concurrently and various other parameters are set via a helios.ini file and the Helios database (the connection information of which is also defined in helios.ini).
Under normal operation, helios.pl will attempt to load the service class specified on the command line and use it to read the contents of the helios.ini file. If successful, it will attempt to connect to the Helios collective database and read the relevant configuration parameters from there. If that is successful, helios.pl will daemonize and start servicing jobs of the specified class.
In debug mode (set HELIOS_DEBUG=1), helios.pl will not disconnect from the terminal, and will output extra debugging information to the screen and the Helios log. It will also enable debug mode on the service class, which may also support the output of extra debugging information.
If the --clear-halt option is specified, helios.pl will attempt to remove a HALT parameter specified in the Helios configuration for the specified service on the current host. This is helpful if you shutdown your Helios service daemon using the Helios::Panoptes Collective Admin view. Note that it will NOT remove a HALT specified globally (where host = '*').
The initial parameters for helios.pl are defined in an INI-style configuration file typically named "helios.ini". The file's location is normally specified by setting the HELIOS_INI environment variable before helios.pl is started. If HELIOS_INI is not set, helios.pl will default to the helios.ini in the current directory, if present.
A helios.ini file normally contains a [global] section with the parameters necessary to connect to the Helios collective database, and any parameters local to the host on which the Helios service is currently running. In addition, each service class may have its own section in helios.ini containing parameters specfic to that service on that host.
[global] dsn=dbi:mysql:host=10.1.0.21;db=helios_db user=helios password=password [LongRunningJobService] master_launch_interval=60 zero_launch_interval=90 [SearchIndex::LoadTestService] OVERDRIVE=1 MAX_WORKERS=3
Datasource name for the Helios parameter database. This will also contain the data structures for the underlying TheSchwartz queuing system.
Database user for the datasource name described above.
Database password for the datasource name described above.
The location where helios.pl should write the PID file for this service instance. The default is "/var/run/helios". The name of the PID file will be a variation on the service class's name.
Set the master_launch_interval to determine how long helios.pl should sleep after launching workers before accessing the database to update its configuration parameters and check for waiting jobs. The default is 1 second which works well for normally short-lived jobs (2 secs or less), but that may be overkill for longer-lived jobs (jobs that run longer than 2 sec). Setting the master_launch_interval option may help prevent needless database traffic for longer running jobs.
Set the zero_launch_interval to determine how long helios.pl should sleep after reaching its MAX_WORKERS limit. The default is 10 sec. If jobs are running long enough that helios.pl is frequently hitting is MAX_WORKERS limit (there are waiting jobs but helios.pl can't launch new workers because previously launched jobs are still running), increasing the zero_launch_interval will reduce needless database traffic.
In addition to helios.ini, certain helios.pl configuration options can be set via the Ctrl Panel in the Helios::Panoptes web interface. These configuration options are read by helios.pl from the HELIOS_PARAMS_TB table in the Helios database. Though the Ctrl Panel is designed mostly to provide a standardized interface for service classes to store and retrieve configuration parameters, certain helios.pl control parameters can be set here as well.
The helios.pl daemon will refresh its configuration parameters from the database after waiting master_launch_interval seconds after launching worker processes. If MAX_WORKERS workers are still running, the helios.pl daemon will wait zero_launch_interval seconds to refresh its configuration parameters and try to launch workers again. If you have long running jobs (jobs that last more than a few seconds), resetting these parameters higher may help smooth processing and reduce needless database traffic. See the HELIOS OPERATION section below for more information.
The Helios system is designed to run one daemon process for each service class per host. However, each service daemon can run more than 1 worker process simultaneously. The upper limit for worker processes is set using the MAX_WORKERS option. The helios.pl default is 1 worker per server. You can provide a different default for a particular class by specifying a value in helios.ini. If active management of available workers is not necessary for a particular service class, this option can be set in helios.ini.
Speed up job processing by allowing workers to service jobs until no more are in the queue. Normally, a worker process will service a single job and then exit (1:1 worker/job ratio). With OVERDRIVE mode enabled, a worker process will keep servicing jobs until it can find no more of that type in the job queue (1:many worker/job ratio). This may help to speed up processing for short-lived jobs, as once a worker process has started, it will stay in memory, continuing to process jobs until there are no more to process. This can also enable on-the-fly caching of resources, as once a worker has loaded a resource or established a database connection for a job it can be held for use by subsequent jobs, potentially reducing database load or disk I/O. Longer running jobs may not benefit as much, however.
Like MAX_WORKERS, if active management of the job processing mode is not necessary, this parameter can be set in helios.ini (though that is discouraged).
If set to 1, a HOLD parameter will cause the helios.pl service daemon to stop launching new workers to process jobs. Currently processing jobs will be allowed to complete, and those worker processes will end normally.
Setting HOLD to 0 will allow normal job processing to resume.
If a HALT parameter is defined, the helios.pl service daemon will start process cleanup and will exit. As with HOLD = 1, jobs currently processing will complete, and the worker processes will exit normally.
Once a helios.pl daemon has HALTED, there is no way to restart it from the Helios::Pantoptes interface. It must be restarted from the command line with a 'helios.pl <serviceclass>' command.
As of Helios 2.22, service daemons can also be shut down in the normal *nix manner of using the 'kill' command to send the daemon a SIGTERM signal.
The time in seconds to allow a worker to run. If you have a problem with workers that get stuck while performing a job (perhaps waiting for a database connection, or some other problem), you can set WORKER_MAX_TTL for that service. The helios.pl program will check to see if workers for that service are running beyond their intended time-to-live, and kill them if they run too long.
After initial setup, the helios.pl daemon will enter a main operation loop where configuration parameters are refreshed, the job queue is checked, and worker processes are launched and cleaned up after. A HOLD = 1 parameter will temporarily cause the loop to pause processing, while a HALT parameter will cause the helios.pl daemon to exit the loop, clean up, and exit.
There are several steps in the helios.pl main operation loop:
The daemonize() function is called to turn the Helios program into a daemon servicing jobs of a particular class. It forks a new process, which disconnects from the launching terminal.
Normally, daemonization would also include setting up signal handling, but the daemonize() function isn't called in debug mode, so signal traps are actually set up in the main program.
The double_clutch() function implements the WORKER_MAX_TTL functionality. If the WORKER_MAX_TTL parameter is set for a service, the service daemon periodically calls double_clutch() to check the workers and clean up any that have run too long. The double_clutch() function waits a certain amount of time (WORKER_MAX_TTL_WAIT_INTERVAL secs), then checks the amount of time each of the running workers has been active. If a worker has been running longer than the service's WORKER_MAX_TTL, the worker is killed (by sending it a SIGKILL signal).
The launch_worker() function launches a new worker process. After the fork() from the main process, the new child process will call launch_worker(). The launch_worker() function will instantiate a new TheSchwartz object, set the object's ability (TheSchwartz term) to the class of the loaded service, and starts the worker on its way by calling the work() method. If the OVERDRIVE run mode is enabled for the service, the work_until_done() method is called instead.
The reaper() function is responsible for cleaning up after dead child processes. It's called when helios.pl receives a SIG_CHLD signal. The function reaps any children with waitpid(), removes the children's PID from the $workers hash of running workers, and re-establishes itself as the signal handler for the next SIG_CHLD signal.
The terminator() function is responsible for shutting down a Helios service instance when helios.pl receives a SIGTERM signal. The shutdown process performs several steps:
Once all these steps are complete, the helios.pl program exits.
Writes a PID file to a location (defaults to /var/run/helios) to track which daemons are running. The file will be named after the service class running, all lowercase, with colons replaced by underscores. For example, the PID file for a service class named 'SearchIndex::LoadTestWorker' will be named "searchindex__loadtestworker.pid". To change the location where the PID file is created, set the pid_path option in helios.ini.
The running_process_check() function should be run before this function to exclusively lock the PID file.
During a clean shutdown, remove_pid_file() is called to delete the PID file associated with the service daemon.
Given the pid_path, check to see if a $pid_file for the loaded service class exists and, if it does, check to see if that process is still running. If the file doesn't exist or it does but isn't running, this function returns 0. If the process is still running, record the error and return the running process's pid to signal that service startup should halt.
If the PID file does not exist, running_process_check() creates it. The PID file is then exclusively locked to prevent other daemons for the same service from writing to it. This helps ensure multiple Helios daemons for same service will not start up at the same time.
The clean_shutdown function is called when helios.pl is intentionally shutdown (setting a HALT parameter in the Helios::Panoptes Ctrl Panel or sending the helios.pl process a TERM signal). It removes the PID file created on startup and unregisters the service instance from the collective database.
The register() function records information about the currently running worker daemon in the database. The register() function is designed to be run every $REGISTRATION_INTERVAL seconds. That way, if a service daemon dies off unexpectedly (without calling unregister()), it can be determined that something has happened to the daemon and it possibly needs to be restarted.
(In reality, register() and unregister() are only necessary to provide a display for Panoptes, to more easily assess system status and facilitate the HALTing of service daemons or HOLDing of job processing.)
The unregister() function removes any record of the currently running daemon from the database. It is called whenever there is a clean shutdown.
Set a HALT parameter for the currently loaded service on the current host to signal to all of the worker processes that they need to exit. This function is used by the terminator() function to safely signal to workers they need to exit when the current job is completed.
If the --clear-halt option is specified on the command line, clear_halt() is called to attempt to clear the HALT parameter in the HELIOS_PARAMS_TB. For safety reasons, it only clears a HALT for the loaded service class AND the specific host helios.pl is running on; it will not clear a global HALT parameter (where the host is specified as '*').
Andrew Johnson, <lajandy at cpan dotorg>
Copyright (C) 2007-9 by CEB Toolbox, Inc.
Portions of this software, where noted, are Copyright (C) 2012-2013 by Logical Helion, LLC.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.8.0 or, at your option, any later version of Perl 5 you may have available.
This software comes with no warranty of any kind.