????? ???????????? wrote: > 2009/11/5 Vladimir Kotal <Vladimir.Kotal at sun.com>: >> ????? ???????????? wrote: >>> ksh93 update 2 landed in build 128 >>> >>> [http://mail.opensolaris.org/pipermail/opensolaris-announce/2009-November/001318.html]. >>> >>> Can you confirm or deny that the update fixed your problem, please? >> With machine bfu'ed with onnv-gate:2009-11-04 (daily bits) and fresh >> {SUNWstc-tetlite,SUNWstc-net-nc} packages I am no longer seeing the behavior >> with extra helper process. > > OK. Would you write a small test case which reproduces the the extra > helper process bug, please? I like to include this in the ksh93 test > suite.
I tried to replicate this behavior in standalone ksh script when I realized nc test suite does not work with ksh93 but was not successful; this worked fine: #!/usr/bin/ksh NC=/usr/bin/nc eval "${NC} -l 4444 &" integer pid="$!" print "pid $pid" The value of the 'pid' variable was the same as the pid of the nc process. There is probably something in CTI-TET environment which triggers this behavior. For the rflag test purpose I will look into what is the actual problem there. v.