{{quegrp -g queue-group-ID -f physical-file-name [-n I/O-buffer-count] [-m retained-message-count] [-w usage-warning-ratio] [-c warning-cancel-percentage]}}

A physical file and the corresponding physical file environment (I/O buffer count, retained message count, etc.) is called a queue group; and the same resources are used by each queue group. The identifier allotted to a queue group is known as a queue group ID. This queue group ID is specified to dynamically add a queue file to the physical file of a given queue group upon a request from MCF. Queue group IDs and physical file names must be unique for all queue groups.

No physical file name must be duplicated with other message queue service definitions.

-g queue-group-ID~<1-8 character identifier>

Specify the queue group ID allotted to the physical file.

MCF uses this identifier to determine which queue groups to allocate to a queue file. Specify this queue group ID if defining the mcfmqgid command with the MCF manager definition.

-f physical-file-name~<path name>

Specify the complete path name as the name of the corresponding physical file. No environment variables can be used.

-n I/O-buffer-count~<unsigned integer> ((2-1024)) <<128>>

Specify the number of input/output buffers for buffer cache use for the corresponding physical file. Buffer caching is performed by allocating several queue files to a physical file. For this the input/out buffers use shared memory. The larger the number of I/O buffers, the smaller will be the number of I/O operations to the disk.

Compute the number of I/O buffers by the following formula:

L = Average message length/Physical file record length x Average number of messages per transaction x Number of transactions concurrently accessing queue files allocated to the same physical file.

N = L + number of records to be overwritten with delay at message validation at intervals of overwriting delay^#

M = L + Length of message to be overwritten with delay/Physical file record length x Number of retained messages x Proportion of messages to be over written with delay^# x Number of queue files allocated to physical file

: Rounded up to the nearest whole integer.

#: The value is collected from the statistics. Specify N or M, whichever may be smaller.

-m retained-message-count~<unsigned integer> ((0-1024)) <<10>>

Specify the number of messages to be retained in each queue file of the queue group so that the user can re-send retained messages with the UAP function dc_mcf_resend. The number of retained messages is the number specified in this operand x the number of queue files.

Messages that the MCF communication service has retrieved from queue files and sent to the other system may be lost before either reaching the other system or in the other system due to a communication or other type of error. If this situation occurs, you may have to resend the messages. The maximum number of messages that you can resend is the value of this operand.

If the number of messages you have sent is greater than the value of this operand, some of the oldest messages cannot be retrieved or resent.

This option can be specified only when the queue group is used as a send queue. If the queue group is used as an input queue, specify 0 in this operand.

-w usage-warning-ratio~<unsigned integer> ((0-95)) <<80>> (Unit: %)

Specify a warning ratio to indicate the amount of a used physical file. Used memory is the memory being used by messages waiting to be read and by retained messages. A warning message is issued if the size of the used physical file (size of the physical file used by messages waiting to be read and by retained messages) exceeds a specified percentage of the physical file size.

-c warning-cancel-percentage~<unsigned integer>((0-95))<<0>> (Unit: %)

For each physical file, specify the percentage for canceling the warning issued for the size of a used physical file (size of the physical file used by messages waiting to be read and by retained messages). If the percentage of the used physical file is lower than the value specified in this operand, warning status is canceled. If the usage warning ratio is exceeded again after the warning status is canceled, a warning message is output. If you specify a value greater than the usage warning ratio for this operand, the system assumes the usage warning ratio.