On Sat, Jun 02, 2012 at 03:17:34PM -0600, Harry G McGavran Jr wrote: >>>I thought you had requested in which 1.7.10s snapshot it broke. >>>So the earliest snapshot I could find for 1.7.10s does have the problem. >>>It works in 1.7.9. >>> >>>As I mentioned earlier in cygwin-inst-20120415.tar.bz2, a 1.7.11s >> >>That should be cygwin-inst-20120220.tar.bz2. >> >>>snapshot it's fixed. That's the snapshot that was current when >>>it was suggested I try the latest snapshot. >>> >>>So, you must be wanting me to check the 1.7.11s snapshots to see >>>if it's fixed in an earlier 1.7.11s snapshot. >>> >>>I'll have to check that out next week when I should have more time. >>> >>> Harry >>> > >I had a little bit of time this afternoon to start on the snapshots >prior to 20120220. I had enough time to try several of these instead >of the first 1.7.10s snapshot I reported on. Since it is necessary >to stop any and all cygwin processes before switching the cygwin1.dll's >each time I try a snapsnot, I ended up going through several iterations >where I remembered to stop everything started by cygrunsrv at startup, >but forgot on occasion to stop gpg-agent. Interestingly, the problem >occurs when I have not stopped gpg-agent. It turns out that I built >gpg2 on cygwin with the resulting gpg-agent at the same time I had >upgraded to 1.7.15. So this problem with 1.7.15 and later is because >exiting mintty does not stop the gpg-agent that is somehow associated >with it and according to an earlier post in this thread, unless >a process associated with mintty is started with setsid, if it >is still running you could have a problem like this.
That is because a subprocess which still has an an open pty associated with it maintains a connection to the originator of the pty. setsid breaks that connection. cgf -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
