On Tue, 5 Feb 2008 12:10:25 +0100, Maarten Slegtenhorst <[EMAIL PROTECTED]> wrote:
>... >When TCP/IP is started, you make changes to SYS1.TCPPARMS (PROFILE) and >try to apply the member using OBEYFILE? > >If so... Then that won't work. Not everything in your PROFILE is >OBEY-able. >... FSVO "won't work". You will get a lot of error messages about devices already started, ports already defined, etc., but (assuming no typos) everything that can be Obeyfiled will be Obeyfiled - just as well as if those statements were moved to a different dataset. Obeyfiling the entire profile definition is a messy but reasonable technique. (If nothing else, it syntax checks the whole profile - something you might want to do before the next time you restart TCP/IP. On the other hand, if you are Obeyfiling in something that may not work, I would want those defs in their own file. Scanning through a flood of gratuitous error messages looking for some important ones can be a real pain. But all this is beside the point in this case. As others have mentioned, TCP/IP complained that it couldn't find the file. It won't complain about the contents of the file until it can read it. Pat O'Keefe ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
