Thank Peter and Nick Jones for your explanation. Only now I could answer.
Thinking about compatibility: when the SMF is written on SMF files (MANx) it is possible to read, with a user program (in assembler, obviously), any records by user's criteria. Now, using SMF Logstream this will not be possible if IBM doesn't give us support with IXGBRWSE. I know installations which have this kind of access on SMF MANx files to get some informations as soon as possible. I think that has good services using this kind of facility. I think IXGBRWSE could be more flexible on z/OS 2.1 and give an option to the caller specifies a parameter about the compression or this "uncompression" could be another available standard service. José ADAUTO Ribeiro De: Peter Relson < [email protected] > Enviada: Segunda-feira, 18 de Novembro de 2013 16:14 Para: [email protected] Assunto: Re: IXGBRWSE - SMF Logstream IXGBRWSE returns via the buffer keyword the exact area that the exploiter requested to be written to the logstream. Thus, it is that exploiter that knows the mapping. I conclude, by the fact that SMF apparently documents only the SMF record itself (and that an SMF record is only part of the logstream record), that ether - SMF does not want you to be reading these records and dealing with them on your own; or - it's an oversight; or - it was thought about but was not thought worthwhile. Is there a business need for reading the logstream that such a program as IFASMFDL does not accommodate? Peter Relson z/OS Core Technology Design ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
