AW: Re: AW: Re: permissions to /bin/sh
>on z/OS the sticky bit on an executable file affects how a spawn() and >_BPX_SHAREAS works. It mainly affects where the loader starts looking for the program to be loaded, namely LPA. Its a performance thing in the first place. Then, for some reason unknown to me, IBM had to decide that such modules must not be loaded as local process. I guess it's an system integrity issue. Therefore it affects spawn(). >Its kind of a hack (mess). In what sense do you think it is a hack? I agree its a mess, in that is inhibits a locally spawned shell. -- Peter Hunkeler -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
AW: Re: AW: Re: permissions to /bin/sh
>1. The problem with IDz was real and solved by change the permissions. To me this is not a solution rather a circumvention to the real problem. No process should recognize the difference in calling, i.e. fork()/exec() or spawn(), a UNIX binary having or not having the sticky bit set. If iDZ is having problems with the shell when the sticky bit is not set, there is something wrong in the iDZ code. It is not using the proper calls / funtions() in some way or the other. -- Peter Hunkeler -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
