Cygwin CVS (or 1.1.8) fixes PostgreSQL postmaster shutdown problem

[Jason Tishler]

PostgreSQL running on 1.1.7 (and earlier) had a minor (but annoying)
postmaster shutdown problem.  If more than one client connected
concurrently to postmaster (i.e., the PostgreSQL backend), then
the postmaster process would not shutdown cleanly.  By "cleanly," I
mean that it would take a random number of "kill -s SIGTERM" attempts
before postmaster would terminate.  Further, the greater the number of
concurrent clients the greater the number of kills before postmaster
would finally exit.

I have been working with Yutaka tanida off and on to try to get to the
bottom of this problem.  See attached for his latest attempt.

Magically the problem seems to be solved in the latest Cygwin CVS.  Did
someone commit a patch that would correct this problem?

Thanks,
Jason

-- 
Jason Tishler
Director, Software Engineering       Phone: +1 (732) 264-8770 x235
Dot Hill Systems Corp.               Fax:   +1 (732) 264-8798
82 Bethany Road, Suite 7             Email: Jason.Tishler@dothill.com
Hazlet, NJ 07730 USA                 WWW:   http://www.dothill.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signalfix.diff
Type: text/x-diff
Size: 909 bytes
Desc: not available
URL: <http://cygwin.com/pipermail/cygwin-developers/attachments/20010205/3a0f63ca/attachment.bin>