So what to do ? Isnt it somewhat overheaded to open a new database connection on every fork, or was it designed for a usage like that ?
BTW : is there someting like a documentation of the DBTech::sapdb modules ? Regards Kai > -----Urspr�ngliche Nachricht----- > Von: Dittmar, Daniel [mailto:[EMAIL PROTECTED] > Gesendet: Donnerstag, 7. August 2003 14:16 > An: '[EMAIL PROTECTED]'; [EMAIL PROTECTED] > Betreff: RE: Perl fork question > > > i have a small perl script checking up port in a network, for > > each ip it > > forks for speedup reasons like this, maximum forks is 20: > [...] > > 2.) is the perl driver safe to use a db-handle before forktime in the > > forked child later (is the same handle in all childs i guess)? > > Inheriting a database handle through fork () is simply a bad idea for SAP > DB. The communication layer checks explicitely on every call that the pid > hasn't changed since the connect. If the pid changed, then the local > communication structures are cleared (without closing the communication). > Linux is an exception, because every thread has it's own pid and we > couldn't find an efficient way to distinguish multiple processes from > multiple threads. > > If multiple clients would be waiting for an answer from the same server > socket, how should they know to whom the answer is actually addressed? > > Daniel Dittmar > > -- > Daniel Dittmar > SAP DB, SAP Labs Berlin > [EMAIL PROTECTED] > http://www.sapdb.org/ > _______________________________________________ sapdb.general mailing list [EMAIL PROTECTED] http://listserv.sap.com/mailman/listinfo/sapdb.general
