All incoming streams may be recorded.
There are two parts to the specification for the filename: a fixed part and a formatted part. The fixed part is identified as the ‘Base Directory’ in the Recording Options page, and the formatted text as Filename formatting. The use of the ‘\’ character (or '/' in linux) is allowed in the filename format, which allows the user to generate a directory structure to suit their needs.
Previous versions of scream enforced a subdirectory containing the stream ID. This is equivalent to adding “A\” to the start of the filename format field.
Typical variations may be to store the data in subdirectories with the day, month, or year. The sample rate or component type may also be used. It is possible to specify more than one level of subdirectory. For example “M\T\DD_HHNNSS” would store in a monthly directory, then by stream ID. The filename would be the start time of the data contained.
The file name is generated in accordance with the file format specifier in the recording setup options, representing the date and time of the start of the time period for that duration. E.g. if two minute files are specified, then the filenames will represent 10,12,14,16 minutes etc. If three minute files are used, the filenames will represent 3,6,9,12 minutes etc.
Recording continues until a write error occurs, or user command terminates recording. If recording fails, Scream can generate a warning email. The three most common reasons for a write error are:
See Also :
GCF Block Specification, Decoding Dates and Times