On Fri, Sep 30, 2005 at 10:09:58PM -0400, Volker Quetschke wrote: >Christopher Faylor wrote: >>The latest snapshot is chock full of developer goodness and needs to >>be tested. >> >>Both Corinna and I spent a lot of time trying to track down the reported >>fork() problems. We didn't fix them* but this snapshot contains better >>fork error reporting. > >Sorry for always being a bit late but testing my favorite testcase >takes a while. Basically we see the same problem as in > <http://cygwin.com/ml/cygwin/2005-09/msg00803.html>. >(This is with 20050929) > >The build hangs in a tcsh command like: > PID PPID PGID WINPID TTY UID STIME COMMAND > 3968 1 3968 3968 con 11290 20:18:06 /usr/bin/bash > 556 3968 556 3436 con 11290 20:18:15 /usr/bin/tcsh > 2880 556 2880 340 con 11290 20:18:51 /usr/bin/perl > 3944 2880 2880 2644 con 11290 12:25:33 > /cygdrive/e/work/OOo/SRC680_m124/solenv/wntmsci10/bin/dmake > 4020 3944 2880 2620 con 11290 12:25:33 /usr/bin/tcsh > 2296 4020 2880 2296 con 11290 12:25:33 /usr/bin/tcsh > 3992 1 3992 3992 con 11290 12:41:26 /usr/bin/bash > 776 3992 776 3368 con 11290 12:41:49 /usr/bin/ps > >And by attaching strace to the hung pid and using "ls /proc/*/fd" >the attached (behind the cygcheck.log) strace was created. > >We tried "kill -s CONT <pid>" this time, that didn't work, the ls did. > > Volker > >P.S.: We (?) will try the next snapshot ASAP, unfortunately this >means in this case on the next weekday (monday).
I wouldn't expect the next snapshot to fix this, unfortunately. I'm still puzzling over where the "open_shared" messages are coming from. Each one repeats 53 times, which is sort of an odd number. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
