FILEMGR Configuration Syntax (UNIX)

For UNIX, the FILEMGR Configuration is used in the UNIX File Collector section of the UNIX Operations of Enterprise Server Management. To make use of the UNIX File Collector functionality the FILEMGR Configuration should be set using the following syntax.

SUBSYS FILEMGR

ADD FILE (<File Name>, <File Name>, ....)
ADD FILEOPEN (<Interval>, <Units>, [<n>])
ADD FILEWHERE (<Where Clause>, [{PERFS|ALLFS}], [{Top|Bottom} <n> by <field>], [<reason>])
ADD DIR (<Directory Name>, <Directory Name>)
ADD DIRWHERE (<Where Clause>, <FS Limiter>, {Top|Bottom} <n> by <field>, <reason>)
DIRSCAN (<Directory Name>, <Directory Name>)
DIRIGNOR (<Directory Name>, <Directory Name>)
INTERVAL (n)
STARTIME (n)
STOPTIME (n)

Syntax Elements

ADD FILE (<File-name>,<File-name>,...)

This parameter is used to specify the files to be monitored. It allows individual files to be specified that will always be seen in the MPFILE and MPDIR records, regardless of the file state. Each entry must include the full path to the file. Where there is more than one file to be included, use a comma (,) as the separator. This entry works independently of the DIRSCAN parameter.

ADD FILEOPEN (<interval>,<units>,[<n>])

Monitors open files.

<Interval>

Sets the time interval for monitoring open files.

<Units>

Specifies the unit of time for the Interval, e.g. minutes, hours.

<n>

The maximum number of open files to monitor.

ADD FILEWHERE (<where-clause>,[<FS limiter>],[{Top|Bottom} <n> by <field>], [<reason>])

This parameter monitors files defined by a Where Clause and optionally allows you to select the top or bottom set of files that meet the 'Where Clause' criteria. Only those files/directories which exist under directories specified in the DIRSCAN parameter will be tested against the respective where clauses.

<Where Clause>

Enter a Where Clause upon which file monitoring is to occur. This consists of standard Where Clause formats.

{PERFS|ALLFS}

Indication of the scope of the Where Clause. PERFS means that the scope is applied per file system and ALLFS means that the scope is applied to all the file systems. Default: ALLFS

[{Top|Bottom} <n> by <field>]

This parameter allows for either the top or bottom number of files that meet the Where Clause condition to be selected. The maximum number of files is nominated in <n> parameter and a specific field to search on can be nominated in the <field> parameter. Using this facility will help to reduce the amount of traffic generated on your network during the file scanning process. If {Top|Bottom} is omitted, no topn clause is applied.

<reason>

Single character field to identify the reason of the Where Clause. This reason flag can be used as a filter to provide tailored reporting for specific applications. If <reason> is omitted, a reason of 'where' is the default.

ADD DIR (<Directory-name>, <Directory-name>)

Directories to scan recursively ( ',' is the separator if more than one directory is to be scanned). This parameter works independently of DIRSCAN, and allows you to specify individual directories that you wish to always see in the MPFILE and MPDIR records respectively, regardless of their state.

ADD DIRWHERE (<where-clause>, <FS limiter>, top|bottom <n> by <field>, <reason>)

This monitors directories defined by a Where Clause and allows you to select the top or bottom set of directories that meet the 'Where Clause' criteria. Only those files/directories which exist under directories specified in the DIRSCAN parameter will be tested against the respective where clauses.

<where-clause>, <FS Limiter >, top|bottom and <reason> fields have the same meaning as for the ADD FILEWHERE parameter.

DIRSCAN (<Directory-name>, <Directory-name>)

This parameter is used to specify the directories that are to be scanned.  Insert a directory path and name. The collector will scan these directories and all their descendants only. DIRIGNOR can be used to ignore specific subdirectories. ADDFILE and ADD DIR can be used to add specific files and directories that are not included in the nominated directories. A comma ',' is the separator if there is more than one directory to scan. The default is to scan all available directories.

DIRIGNOR (<Directory-name>, <Directory-name>)

This parameter is used to specify directories that are to be ignored during scanning. Insert a directory path and name. The specified directories and all their descendants will be ignored during recursive disk scanning. A comma ',' is the separator if there is more than one directory to ignore. The default is to scan all available directories.

INTERVAL (n)

Minutes between the end of the recursive scanning and the start of the next. Scanning will only take place between the STARTIME and STOPTIME.

STARTIME (n)

Start time for the disk scanning in hours of the day, e.g. enter 18 for 1800 being 6pm. The default is 0.

STOPTIME (n)

Stop time for doing the disk scanning collection in hours of the day, e.g. enter 18 for 1800 being 6pm. The default is 6.

If scanning is still running when the STOPTIME is reached, scanning will cease and a message will be logged to the WVLOG file to indicate that the scan interval is not sufficient to complete scanning. In addition, the File Collector will remember that the last scan was not completed and will send an unsolicited message to any online Displays indicating that the sample data may be inaccurate.

If you have set a STARTIME and STOPTIME and selected to use the 'Buffered' mode (i.e. Mode 1), the program priority will be NORMAL during these times. Outside the time span the program priority will be LOW.

Provide feedback on this article