On Mon, May 04, 2009 at 10:54:55AM -0500, Thrall, Bryan wrote: >Christopher Faylor wrote on Monday, May 04, 2009 10:44 AM: >> On Mon, May 04, 2009 at 10:35:46AM -0500, Thrall, Bryan wrote: >>> Christopher Faylor wrote on Monday, May 04, 2009 10:18 AM: >>> >>>> On Mon, May 04, 2009 at 03:11:02PM +0000, Greg Chicares wrote: >>>>> On 2009-05-04 14:44Z, Thrall, Bryan wrote: >>>>>> Corinna Vinschen wrote on Monday, May 04, 2009 3:25 AM: >>>>>>> On Apr 28 11:47, Julio Costa wrote: >>>>>>>> Bug in cygcheck? Or bug in user? :) >>>>>>> >>>>>>> Bug in user as far as the position of the -v option is concerned. >>>>>> >>>>>> I can reproduce it; some more details: >>>>>> >>>>>> - 'cygcheck /bin/ash.exe -v' returns 127 >>>>>> - it doesn't happen if run from cmd.exe >>>>>> - '/bin/ash.exe -v' from inside bash works just fine >>>>>> >>>>>> Isn't there a feature of Cygwin where it tries 5 or 6 times to >restart >>>>>> Windows apps (such as cygcheck) if they fail in a certain way? >That >>>>>> could explain the repeated output (i.e. cygcheck is trying to do >the >>>>>> documented 'cygcheck [PROGRAM] -v' thing but there's a bug), but >my >>>>>> search fu wasn't able to bring up any results. >>>>> >>>>> Search for 'proc_retry' in the User's Guide. >>>>> >>>>> Try setting >>>>> CYGWIN=proc_retry:1 >>>>> to see whether that's what's happening in this case. >>> >>> Thanks, Greg, for the pointer! >>> >>>> If it is, then cygcheck would be exiting with a bad status and bash >>>> would be causing cygcheck to restart. You wouldn't see the problem >>>> from the windows command line. >>> >>> Aren't those the exact symptoms I described? >> >> The "exact symptoms"? No, in fact it was not. I was clarifying what >to >> look for but I guess I should have said "Please try this from the >> windows command line". > >Sorry, I did say: >>>>>> - 'cygcheck /bin/ash.exe -v' returns 127 >>>>>> - it doesn't happen if run from cmd.exe
Sorry I missed that. >But I can be more exact: >If I run 'cygcheck /bin/ash.exe -v' from inside cmd.exe, I only get one >copy of the output. >I didn't think to check this before, but %ERRORLEVEL% after running that >is -1073741819 (I think that's the normal way to check return values in >cmd.exe?). The windows error code is what I was looking for but again I wasn't being very clear. That is a STATUS_ACCESS_VIOLATION, which would be a cygcheck bug. Wish I could duplicate it. 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/
