I have never launch an app without using a servlet container like tomcat or JavaMonitor so this was not an issue for me. Deploying the war on tomcat does not required and webobjects setup, and does not use NEXT_ROOT. Now that I understand your issue, I see your problem, just was never a problem for me in my linux deployments
On Thu, Jun 28, 2012 at 9:05 AM, Gavin Eadie <[email protected]> wrote: > This happens in "direct connect" mode, at least, which is what I'm > using to gather data. It's not being 'deployed' as such, just run as > "xxx.woa/xxx". > > I suspect you, like I, have had NEXT_ROOT defined on all the WO > machines we use for years -- I certainly haven't had to bother with > that variable for a long time -- but there are essentially virgin > machines with no prior experience of WebObjects. > > It's clear in the 'launch script' how this happens: > > if [ "${PLATFORM_TYPE}" = "Other" ] > then > if [ "${NEXT_ROOT}" = "" ] > then > echo ${SCRIPT_NAME}: NEXT_ROOT environment variable is not > set! Terminating. > exit 1 > fi > fi > > I can tell them all to create NEXT_ROOT with non-null value but this > offends my sense of correctness ... > > On 6/25/12, Ron Lift <[email protected]> wrote: > > How are the applications being deployed? I have deployed using tomcat and > > JavaMonitor on RedHat without any problems, just embeding the frameworks > > and the jars. > > > > On Mon, Jun 25, 2012 at 12:56 PM, Gavin Eadie > > <[email protected]>wrote: > > > >> If you build an application with all its frameworks embedded, the > >> environment variable NEXT_ROOT is never used. If you then deploy this > >> app on a non-Windows/non-MacOSX system, the launch script > >> "xxx.woa/xxx" will terminate because NEXT_ROOT is not defined, and the > >> app will not launch. I suspect that embedding and linux deploys are > >> becoming more common so, while this might have been an unusual > >> circumstance years ago, it isn't unusual today. > >> > >> It is, of course, easy to define the variable and dodge this issue, > >> but I find myself in a situation where this is less desirable. I'm > >> sending out an 'all-frameworks-embedded' app to a few people to run in > >> direct connect mode and one-by-one they've tripped over > >> > >> NEXT_ROOT environment variable is not set! Terminating. > >> > >> I could fix this by dropping an export NEXT_ROOT="" into the top of > >> the launch script, but I don't like changing standard files (somebody > >> downstream will get confused). I could change the WOLips build.xml > >> file or (gasp) the <woapplication> ant task but that's even less > >> attractive. > >> > >> The mail list archive don't have any discussion of this, and I'm > >> wondering if this is something that has been resolved in a way that is > >> widely accepted by this community as the "right way"? > >> _______________________________________________ > >> Do not post admin requests to the list. They will be ignored. > >> Webobjects-dev mailing list ([email protected]) > >> Help/Unsubscribe/Update your Subscription: > >> > https://lists.apple.com/mailman/options/webobjects-dev/rpgile4%40gmail.com > >> > >> This email sent to [email protected] > >> > > >
_______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [email protected]
