I had the same thought Gord. I already tried changing the default "LOOKUP DNS LOCAL" to "LOOKUP DNS" and that didn't change anything.
On Tue, Jan 22, 2019 at 10:44 PM Gord Tomlin < gt.ibm.li...@actionsoftware.com> wrote: > On 2019-01-22 18:42, Kirk Wolf wrote: > > I've noticed that one *some* z/OS systems that the first call to the > > resolver by a Unix process takes an extra 1/2 second or more. > > Does anyone have any clues as to why this would be? > > > > I have a tiny test program that demonstrates. It calls > > getaddrinfo(127.0.0.1) twice. > > > > I've highlighted the lines in the resolver trace output below that wrap > the > > lost time (577 ms). > > The time is lost regardless of whether I run this without the resolver > > trace. > > On another z/OS system, this same section (before res_init Started) takes > > less than 2 ms. > > Also, if I run the program multiple times from the same shell with > > _BPX_SHAREAS=YES, each run demonstrates the problem. > > So the lost time is per-process and not per-address space. > > > > Thanks, > > > > Kirk Wolf > > Dovetailed Technologies > > http://dovetail.com > > Do you have a LOOKUP statement in your TCPIP.TCPIP.DATA (or equivalent), > and if so what is the lookup sequence? > > -- > > Regards, Gord Tomlin > Action Software International > (a division of Mazda Computer Corporation) > Tel: (905) 470-7113, Fax: (905) 470-6507 > Support: https://actionsoftware.com/support/ > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
