I have noticed this problem with rsh in a number of cases. I saw reference to the same/similar problem in the MLA to which Corinna responded that it would be fixed in inetutils-1.3.2-10. I am using inetutils-1.3.2-14. Here are some examples of what seems like odd behavior to me:
rsh hostname "ls -l /cygdrive/c/foo" (returns nothing) rsh hostname "cd /cygdrive/c/foo; ls -l" (returns a full or partial directory listing) rsh hostname "cd /cygdrive/c/foo; ls -l; sleep 3" (returns the correct directory listing and sleeps) rsh hostname "ls -l /cygdrive/c/foo; sleep 3" (returns the correct directory listing and sleeps) rsh hostname "ls -l /non/existent/dir; sleep 3" (returns nothing and sleeps) rsh hostname "cat localdir/foo.c" (cats the whole file) rsh hostname "cd /cygdrive/c/foo; cat foo.c" (cats the whole or partial file, possibly dependent on length?) rsh hostname "cat /cygdrive/c/foo.c; sleep 3" (cats the whole file and sleeps) It looks like rsh doesn't flush the buffer before returning. There are times when I try an ls or cat, and I get my prompt back at a random spot in the file/listing. And I can't get any errors to come back, such as "No such file or directory" when I try to ls a non-existent directory. I am guessing that this is an rsh problem, as opposed to rshd. Most of these tests return the correct output when I try to rsh to hostname running cygwin inetd from a solaris machine (except that I still get no errors when a file/dir does not exist). I am not very experienced with gdb, but if anyone can give me pointers to debug further they would be greatly appreciated. TIA, Peter -- Your mouse has moved. Windows NT must be restarted for the change to take effect. Reboot now? [OK] -- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
