Skip to content

Configuration of DPF Server - main.customer.xml

??? Todo: Was davon läuft wirklich noch über die main.customer.xml und was über den Consul? Wird die main.customer.xml im laufenden Betrieb genutzt oder nur zur Initialisierung? Was ist bei Änderungen?

This chapter describes the configuration settings for:

-> Erweitern und anpassen an heutige main.customer.xml, siehe Master Version 3.3.1: M:\dpf_3.3.1\server\dpf\templates\main.customer.xml.tpl -> Teile raussuchen, und dann mit Entwicklern abklären

Hint - main.customer.xml

Customize the settings in server\dpf\templates\main.customer.xml to prevent the settings from being overwritten when updating the DPF system.

File Format

The following elements of the XML configuration files are predetermined:

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
    ...
    </config>
</DPF>

Each element possesses predetermined keywords which will be evaluated internally by the DPF system. In addition to the predetermined keywords, you may configure any keywords.

Hint - upper/lower case

Note that upper and lower case are considered when reading the configuration files! You must use the element names and keywords in the notation given in this documentation!

Server

For configuring the server for the DPF system and the servers of the remote <nodenames> working unit, the following element is available in the <nodesname> element:

In the <nodenames><node> element, you configure the names of the servers for the DPF systems and the remote working units. The following attributes are available in the <nodenames><node> element:

NAME specifies the name of the server of the DPF system.

In the <nodenames><node><variable> element, you can configure any servers of the remote working units as variable variable. Then, you use this variable in the configuration of the working units.

In the <nodenames><node><DEFAULTNODE> element, you specify the server where all working units are to be executed if no server has been specified in the configuration.

Example - simple remote system

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
    ...
        <nodenames>
            <node NAME="dpf01">
                <REMOTENODES>
                    dpf001 dpf006
                </REMOTENODES>
                <DEFAULTNODE>
                    %DB_HOST%
                </DEFAULTNODE>
            </node>
        </nodenames>
    ...
    </config>
</DPF>

Example - complex remote system with different CAD converters, test and productive systems

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
    ...
        <nodenames>
            <node NAME="dpftst01">
                <ACADNODES>
                    dpftst001
                </ACADNODES>
                <PROE_NODES>
                    dpftst003
                </PROE_NODES>
                <OFFICENODES>
                    dpftst004 dpftst005
                </OFFICENODES>
                <DEFAULTNODE>
                    %DB_HOST%
                </DEFAULTNODE>
            </node>
            <node NAME="dpf01">
                <ACADNODES>
                    dpf001
                </ACADNODES>
                <PROE_NODES>
                    dpf003
                </PROE_NODES>
                <OFFICENODES>
                    dpf004 dpf005
                </OFFICENODES>
                <DEFAULTNODE>
                    %DB_HOST%
                </DEFAULTNODE>
            </node>
        </nodenames>
    ...
    </config>
</DPF>

DPF Gate

For configuring DPF Gate, the following keywords are available in the <dpfgate> element:

ARCHIV

ARCHIV specifies the name of the directory where DPF Gate saves the jobs that were processed without errors.

Available values: String

  • <dir>

    Successful jobs will be saved into the <dir> directory relatively to the process-specific subdirectory.

  • ""

    Successful jobs will not be saved.

Default: ""

ERROR

ERROR specifies the name of the directory where DPF Gate saves the erroneous jobs.

Available values: String

  • <dir>

    Erroneous jobs will be saved into the dir directory relatively to the process-specific subdirectory.

  • ""

    Erroneous jobs will not be saved.

Default: error

LOGLEVEL (Gate)

LOGLEVEL specifies the log level.

Available values: Enumeration

  • LOG_OFF or OFF

    No log messages

  • LOG_RUN or R

    Only start and stop messages; the identifier of these messages is [R].

  • LOG_FATAL or F

    LOG_RUN and fatal error messages; the identifier of these messages is [F].

  • LOG_ERROR or E

    LOG_FATAL and all errors; the identifier of these messages is [E].

  • LOG_WARN or W

    LOG_ERROR and all warnings; the identifier of these messages is [W].

  • LOG_INFO or I

    LOG_WARN and info messages; the identifier of the messages is [I].

  • LOG_DEBUG or D

    LOG_INFO and debug messages; the identifier of these messages is [D].

  • LOG_TRACE or T

    LOG_DEBUG and trace messages; the identifier of these messages is [T].

Default: LOG_INFO

Large file

With LOG_TRACE, the log files become very large and, therefore, slow down the system very much. For this reason, activate this log level only very selectively and deactivate it again as soon as possible!

LOGSIZE (Gate)

LOGSIZE specifies the maximum size of the DPF Gate log file in KB. When reaching the specified size, the log file will be saved with the extension .old and a new log file will be started. If a file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

RDY

RDY specifies the extension of the files triggering the processing by DPF Gate.

Available values: String

  • <ext1> <ext2> ... <extn>

    As soon as a file with one of the specified extensions is stored in the directory, processing by DPF Gate will start.

Default: rdy

SGATE

SGATE specifies the name of the directory where the process-specific subdirectories containing the jobs are located.

Available values: String

  • <dir>

    The process-specific subdirectories are expected in the <dir> directory.

Default: %PLSDATA%\io\dpfgate

SLEEP

SLEEP specifies the waiting time in seconds after that DPF Gate checks the directory again.

Available values: Integer

Default: 4

Example (Gate)

Example configuration of DPF Gate

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <dpfgate
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            SLEEP="4"
            RDY="rdy"
            SGATE="%PLSDATA%/io/dpfgate"
            ERROR="error"
            ARCHIV=""
        />
        ...
    </config>
</DPF>

DPF Gate Directories

In the <directories><directory> element, you can configure the names of the DPF Gate directories for each process. The following attributes are available in the <directories><directory> element:

  • name specifies the name of the directory where the jobs are searched for.
  • workflow specifies the name of the process to that the jobs in the directory will be assigned.

jobParameters

In the <directories><directory><jobParameters> element, you can configure any job parameters for each process.

The following attributes are available in the <directories><directory><jobParameters><jobParameter> element:

  • name specifies the name of the parameter for the jobs. name
  • value specifies the values of the parameter for the jobs.

Example (Directories)

Example configuration of DPF Gate directories

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <directories>
            <directory name="tiff2pdf" workflow="dpf4view.prepare_single">
                <jobParameters>
                    <jobParameter name="CURRFILETYPE" value="TIFF" />
                    <jobParameter name="JobName" value="Gate tiff2pdf" />
                    <jobParameter name="JobUserName" value="tiff2pdf Test" />
                    <jobParameter name="DPF4C_MODIFY_STAMP" value="Y" />
                    <jobParameter name="DPF4C_MODIFY_STAMP_N_0" value="Stamp 0" />
                </jobParameters>
                </directory>
                <directory name="gif2pdf" workflow="dpf4view.prepare_single">
                <jobParameters>
                    <jobParameter name="CURRFILETYPE" value="GIF" />
                    <jobParameter name="JobName" value="Gate gif2pdf" />
                    <jobParameter name="JobUserName" value="gif2pdf Test" />
                    <jobParameter name="DPF4C_MODIFY_STAMP" value="Y" />
                    <jobParameter name="DPF4C_MODIFY_STAMP_N_0" value="Stamp 0" />
                </jobParameters>
            </directory>
        </directories>
        ...
    </config>
</DPF>

Literature - DPF Gate configuration

Refer to Installation and Configuration of DPF Gate.

DPF Job Cleaner

DPF Job Cleaner deletes the finished jobs from the job files’ directory and the MongoDB database. The jobs are deleted in two steps:

  • In the first step, the job files are deleted.
  • In the second step, the complete job directory is deleted that means the information about the job.

For configuring DPF Job Cleaner, the following keywords are available in the <jobcleaner> element:

DELETEJOB (Job Cleaner)

DELETEJOB specifies the deletion time of the job files.

Available values: String

  • atonce

    The job files will be deleted right after the processing.

  • never

    The job files will never be deleted.

  • <week> W <day> D <hour> H <minute> M <second> S

    The job files will be deleted after reaching the specified time.

Default: never

Hint - evaluation

The keyword is only evaluated if it is not specified with the job.

ERR_DELETEJOB

ERR_DELETEJOB specifies the deletion time of the job files of the erroneous jobs.

Available values: String

  • atonce

    The job files of the erroneous jobs will be deleted right after the processing.

  • never

    The job files of the erroneous jobs will never be deleted.

  • <week> W <day> D <hour> H <minute> M <second> S

    The job files of the erroneous jobs will be deleted after reaching the specified time.

Default: never

Hint - evaluation

The keyword is only evaluated if it is not specified with the job.

FORGETJOB

FORGETJOB specifies the deletion time of the job directory.

Available values: String

  • atonce

    The job directory will be deleted right after the processing.

  • never

    The job directory will never be deleted.

  • <week> W <day> D <hour> H <minute> M <second> S

    The job directory will be deleted after reaching the specified time.

Default: never

Hint - evaluation

The keyword is only evaluated if it is not specified with the job.

INSTANCES (Job Cleaner)

INSTANCES specifies the number of processes for DPF Job Cleaner.

Available values: Integer

Default: 1

KEEPUNTILFORGET

KEEPUNTILFORGET specifies the name of the job files kept until the job directory will be deleted. Wildcards are not possible here.

Available values: String

  • <file1> <file2> ... <filen>

    The <file1> <file2> ... <filen> files are kept.

Default: job.dat joborig.dat job.log ORIGFILENAME

Hint - evaluation

The keyword is only evaluated if it is not specified with the job.

LOGLEVEL (Job Cleaner)

LOGLEVEL specifies the log level.

Available values: Enumeration

  • LOG_OFF or OFF

    No log messages

  • LOG_RUN or R

    Only start and stop messages; the identifier of these messages is [R].

  • LOG_FATAL or F

    LOG_RUN and fatal error messages; the identifier of these messages is [F].

  • LOG_ERROR or E

    LOG_FATAL and all errors; the identifier of these messages is [E].

  • LOG_WARN or W

    LOG_ERROR and all warnings; the identifier of these messages is [W].

  • LOG_INFO or I

    LOG_WARN and info messages; the identifier of the messages is [I].

  • LOG_DEBUG or D

    LOG_INFO and debug messages; the identifier of these messages is [D].

  • LOG_TRACE or T

    LOG_DEBUG and trace messages; the identifier of these messages is [T].

Default: LOG_INFO

Large file

With LOG_TRACE, the log files become very large and, therefore, slow down the system very much. For this reason, activate this log level only very selectively and deactivate it again as soon as possible!

LOGSIZE (Job Cleaner)

LOGSIZE specifies the maximum size of the DPF Job Cleaner log file in KB. When reaching the specified size, the log file will be saved with the extension .old and a new log file will be started. If a file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

RETRY_FORGETJOB

RETRY_FORGETJOB specifies the time after that DPF Job Cleaner retries deleting the job information when an error occurred. ??? Todo: Überprüfen, ob das so stimmt

Available values: String

  • <week> W <day> D <hour> H <minute> M <second> S

    DPF Job Cleaner retries deleting the job information after reaching the specified time.

Default: 1m

RETRY_DELETEJOB

RETRY_DELETEJOB specifies the time after that DPF Job Cleaner retries deleting the job files when an error occurred. ??? Todo: Überprüfen, ob das so stimmt

Available values: String

  • <week> W <day> D <hour> H <minute> M <second> S

    DPF Job Cleaner retries deleting the job files after reaching the specified time.

Default: 1m

SLEEP_MILLI_SECONDS (Job Cleaner)

SLEEP_MILLI_SECONDS specifies the number of milliseconds after that DPF Job Cleaner asks the MongoDB database if there is something to do. ??? Todo: Überprüfen, ob das so stimmt? Oder kommt da NATS dazwischen?

Available values: Integer

Default: 250

VERSION

VERSION specifies the version of DPF Job Cleaner. ??? Todo: Specifies? Or Outputs?

Example (Job Cleaner)

Example configuration of DPF Job Cleaner

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <jobcleaner
            VERSION="$Id: main.xml ... Exp $"
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            INSTANCES="1"
            SLEEP_MILLI_SECONDS="250"
            DELETEJOB="3h"
            ERR_DELETEJOB="4d"
            FORGETJOB="2d"
            RETRY_FORGETJOB="1m"
            KEEPUNTILFORGET="job.dat joborig.dat job.log
            ORIGFILENAME"
        </jobcleaner>
        ...
    </config>
</DPF>

DPF Process Manager

For configuring DPF Process Manager, the following keywords are available in the element:

INSTANCES (Process Manager)

INSTANCES specifies the number of processes for DPF Process Manager.

Available values: Integer

Default: 1

LOGLEVEL (Process Manager)

LOGLEVEL specifies the log level.

Available values: Enumeration

  • LOG_OFF or OFF

    No log messages

  • LOG_RUN or R

    Only start and stop messages; the identifier of these messages is [R].

  • LOG_FATAL or F

    LOG_RUN and fatal error messages; the identifier of these messages is [F].

  • LOG_ERROR or E

    LOG_FATAL and all errors; the identifier of these messages is [E].

  • LOG_WARN or W

    LOG_ERROR and all warnings; the identifier of these messages is [W].

  • LOG_INFO or I

    LOG_WARN and info messages; the identifier of the messages is [I].

  • LOG_DEBUG or D

    LOG_INFO and debug messages; the identifier of these messages is [D].

  • LOG_TRACE or T

    LOG_DEBUG and trace messages; the identifier of these messages is [T].

Default: LOG_INFO

Large file

With LOG_TRACE, the log files become very large and, therefore, slow down the system very much. For this reason, activate this log level only very selectively and deactivate it again as soon as possible!

LOGSIZE (Process Manager)

LOGSIZE specifies the maximum size of the DPF Process Manager log file in KB. When reaching the specified size, the log file will be saved with the extension .old and a new log file will be started. If a file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

NODENAMES (Process Manager)

NODENAMES specifies the servers where DPF Process Manager is started.

Available values: String

  • <server1> <server2> ... <servern>

    List of servers where DPF Process Manager is started

  • local`host

    DPF Process Manager is started on every server where sysstart is called.

Default: localhost

SLEEP_MILLI_SECONDS (Process Manager)

SLEEP_MILLI_SECONDS specifies the number of milliseconds after that DPF Process Manager asks the MongoDB database if there is something to do. ??? Todo: Überprüfen, ob das so stimmt? Oder kommt da NATS dazwischen?

Available values: Integer

Default: 100

THREADS (Process Manager)

THREADS specifies the number of threads for DPF Process Manager.

Available values: Integer

Default: 1

WARN_BEFORE_LICENSE_EXPIRES

WARN_BEFORE_LICENSE_EXPIRES specifies the number of days and weeks as of which a warning will be output before the license will expire.

Available values: String

  • <week> W <day> D

    As of the specified number of <days> and <weeks>, a warning is output before the license will expire.

Default: 14d

Example (Process Manager)

Example configuration of DPF Process Manager

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <processmanager
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            THREADS="1"
            INSTANCES="1"
            WARN_BEFORE_LICENSE_EXPIRES="100"
            NODENAMES="localhost"
            SLEEP_MILLI_SECONDS="100"
        />
        ...
    </config>
</DPF>

DPF Watchdog

For configuring DPF Watchdog, the following keywords are available in the element:

CHECK_INTERVAL

CHECK_INTERVAL specifies a factor for the time period in SLEEP_MILLI_SECONDS. After this time period, DPF Watchdog checks if the processes are still running.

Available values: Integer

Default: 100

KILL_COMMAND

KILL_COMMAND specifies the command for stopping the processes.

Available values: String

Default: nkill -sKILL -cr -no -pid %PID%

LOGLEVEL (Watchdog)

LOGLEVEL specifies the log level.

Available values: Enumeration

  • LOG_OFF or OFF

    No log messages

  • LOG_RUN or R

    Only start and stop messages; the identifier of these messages is [R].

  • LOG_FATAL or F

    LOG_RUN and fatal error messages; the identifier of these messages is [F].

  • LOG_ERROR or E

    LOG_FATAL and all errors; the identifier of these messages is [E].

  • LOG_WARN or W

    LOG_ERROR and all warnings; the identifier of these messages is [W].

  • LOG_INFO or I

    LOG_WARN and info messages; the identifier of the messages is [I].

  • LOG_DEBUG or D

    LOG_INFO and debug messages; the identifier of these messages is [D].

  • LOG_TRACE or T

    LOG_DEBUG and trace messages; the identifier of these messages is [T].

Default: LOG_INFO

Large file

With LOG_TRACE, the log files become very large and, therefore, slow down the system very much. For this reason, activate this log level only very selectively and deactivate it again as soon as possible!

LOGSIZE (Watchdog)

LOGSIZE specifies the maximum size of the DPF Watchdog log file in KB. When reaching the specified size, the log file will be saved with the extension .old and a new log file will be started. If a file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

RESTART_COMMAND

RESTART_COMMAND specifies the command for restarting the processes.

Available values: String

Default: %DPFSRV%\dpfstart.pl

RESTART_HANG_COMMAND

RESTART_HANG_COMMAND specifies the command for restarting the processes after DPF Watchdog detected a missing component or a crash.

Available values: String

Default: %DPFSRV%\dpfrestart.pl

SLEEP_MILLI_SECONDS (Watchdog)

SLEEP_MILLI_SECONDS specifies the number of milliseconds after that DPF Watchdog asks the MongoDB database if there is something to do. ??? Todo: Stimmt das so?

Available values: Integer

Default: 250

WARN_BELOW_DISK_SPACE

WARN_BELOW_DISK_SPACE specifies the size of free disc space in GB below that DPF Watchdog outputs a warning into the log file, watchdog1.log.

Available values: Float

Default: 2.0

WARN_BELOW_MEMORY

WARN_BELOW_MEMORY specifies the size of free memory space in GB below that DPF Watchdog outputs a warning into the log file, watchdog1.log.

Available values: Float

Default: 0.25

Example (Watchdog)

Example configuration of DPF Watchdog

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <watchdog
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            SLEEP_MILLI_SECONDS="250"
            CHECK_INTERVAL="100"
            RESTART_COMMAND="%DPFSRV%\dpfstart.pl"
            RESTART_HANG_COMMAND="%DPFSRV%\dpfrestart.pl"
            KILL_COMMAND="nkill -sKILL -cr -no -pid %PID%"
            WARN_BELOW_DISK_SPACE="0.2"
            WARN_BELOW_MEMORY="0.5"
        />
        ...
    </config>
</DPF>

DPF Working Unit Manager

For configuring DPF Working Unit Manager, the following keywords are available in the element:

LOGLEVEL (WU Manager)

LOGLEVEL specifies the log level.

Available values: Enumeration

  • LOG_OFF or OFF

    No log messages

  • LOG_RUN or R

    Only start and stop messages; the identifier of these messages is [R].

  • LOG_FATAL or F

    LOG_RUN and fatal error messages; the identifier of these messages is [F].

  • LOG_ERROR or E

    LOG_FATAL and all errors; the identifier of these messages is [E].

  • LOG_WARN or W

    LOG_ERROR and all warnings; the identifier of these messages is [W].

  • LOG_INFO or I

    LOG_WARN and info messages; the identifier of the messages is [I].

  • LOG_DEBUG or D

    LOG_INFO and debug messages; the identifier of these messages is [D].

  • LOG_TRACE or T

    LOG_DEBUG and trace messages; the identifier of these messages is [T].

Default: LOG_INFO

Large file

With LOG_TRACE, the log files become very large and, therefore, slow down the system very much. For this reason, activate this log level only very selectively and deactivate it again as soon as possible!

LOGSIZE (WU Manager)

LOGSIZE specifies the maximum size of the DPF Working Unit Manager log file in KB. When reaching the specified size, the log file will be saved with the extension .old and a new log file will be started. If a file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

NODENAMES (WU Manager)

NODENAMES specifies the servers where DPF Working Unit Manager is started.

Available values: String

  • <server1> <server2> ... <servern>

    List of servers where DPF Working Unit Manager is started

  • localhost

    DPF Working Unit Manager is started on every server where sysstart is called.

Default: localhost

SLEEP_MILLI_SECONDS (WU Manager)

SLEEP_MILLI_SECONDS specifies the number of milliseconds after that DPF Working Unit Manager asks the MongoDB database if there is something to do. ??? Todo: Stimmt das so?

Available values: Integer

Default: 250

THREADS (WU Manager)

THREADS specifies the number of threads for DPF Working Unit Manager.

Available values: Integer

Default: 10

Example (WU Manager)

Example configuration of DPF Working Unit Manager

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <workingunit
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            THREADS="10"
            NODENAMES="localhost"
            SLEEP_MILLI_SECONDS="100"
        />
        ...
    </config>
</DPF>

Accounting

In the accounting, values are logged which then for example can be evaluated in a statistic. For each job, a line containing the values of the configured parameters are written into the accounting file.

For configuring the accounting, the following keywords are available in the <accounting> element:

LOGFORMAT

LOGFORMAT specifies the text and parameters enclosed by % that are written into the accounting file for each job. Here, you can specify any text and all parameters that are set at the end of the job. Unless the parameter is set, the field remains empty.

Available values: String

Default: %JOB_TIMESTAMP_BEGIN%;%JOB_TIMESTAMP_END%;%JOBID%;%WFNAME%;%ORIGUSERNAME%;%ORIGHOSTNAME%;%JOBSTATUS%

LOGLEVEL (Accounting)

LOGLEVEL specifies here if the line is written into the accounting file.

Available values: Enumeration

  • LOG_OFF or OFF

    Nothing is written into the accounting file.

  • LOG_INFO or I

    The line is written into the accounting file.

Default: LOG_INFO

LOGNAME

LOGNAME specifies the base name of the accounting file.

Available values: String

  • <basename>

    The <basename>.log file will be created in the log directory, data\log.

Default: account

LOGSIZE (Accounting)

LOGSIZE specifies the size of the accounting file in KB. When reaching the specified size, the accounting file will be saved with the extension .old and a new accounting file will be started. If a saved accounting file with the .old extension already exists, it will be overwritten.

Available values: Integer

Default: 1000

Example (Accounting)

Example configuration of accounting

<?xml version="1.0" encoding="ISO-8859-1"?>
<DPF>
    <config>
        ...
        <accounting
            LOGNAME="account"
            LOGLEVEL="LOG_INFO"
            LOGSIZE="1000"
            LOGFORMAT="%JOB_TIMESTAMP_BEGIN%;%JOB_TIMESTAMP_END%;%JOBID%;%WFNAME%;%ORIGUSERNAME%;%ORIGHOSTNAME%;%JOBSTATUS%"
        />
        ...
    </config>
</DPF>
Back to top