On Tue, Oct 14, 2008 at 6:41 PM, Dave Korn wrote: > Thorsten Kampe wrote on 14 October 2008 21:11: > >> * Z W (Tue, 14 Oct 2008 11:59:07 -0700) >>> I'm have a shell script, test.sh below to run with cygwin in a Windows >>> box. >>> >>> #!/usr/bin/sh >>> >>> ssh [EMAIL PROTECTED] 'export MAX_MS=100; export OFFSET_MS=89900; export >>> THREADS=4; export RAMP=1; export LOOPS=2; echo load01 ; cd >>> /cygdrive/c/apps/bin ; pwd ; nohup ./start.sh \& ; ps -efW | grep >>> java ; exit ' >> >> That's not a shell script - it's just a single command. > > Uh, it's a shell script *containing* a single ssh command. Note the > shebang. Also note the mention of the filename "test.sh", which is not to be > confused with the script "start.sh" on the remote machine.
I'm with you here... >> The multiple export statements are superfluous. > > Surely not, if you want the definitions to be visible in the sub-shell that > executes "start.sh"? I think Thorsten meant that a single "export" with multiple var=val pairs would have sufficed here. >> The "cd" looks superfluous, too. > > Surely not, unless /cygdrive/c/apps/bin is the qa user's home dir? Definitely not superfluous, since ./start.sh is in that dir.. >> Also the "pwd". > > Surely not, if it's a bit of debugging trace that lets ZW know that the > directory exists and so the cd command succeeded and there must be something > else wrong if the script doesn't execute? Yep. >> Also the "exit". > > Ok, I'm with you there :-) But I still think one of us needs more coffee... Yep Yep. ~Matt -- 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/
