Hi guys,
I’m having a little issue here.
I’m installing a DBServer on Linux for tests and the issue is that if I start it from command line (run.sh) it works fine, clients can do queries and stuff. However if I start it via SCI then seems it can’t get the Environment variables I declared on bashrc (Oracle ones) and can’t make it work, getting error code 1127. Any idea on how to solve this? Seems to be how LCA sends the start command but can’t figure out how it does the execution.
The user account under which is the LCA running has these environment variables in his profile?Try to log in under this account and look at the env or try to restart LCA, I remember that in a past I have registered similiar issue, when the LCA was badly cached the profile.
LCA is running under root, the same account I’m using to start the DBserver manually, the Cfg DBserver does work fine but of course a starting it manually…really don’t understand what am I missing…
Well created a custom script inside /etc/profile.d/ (gcti.sh) and still didn’t work…
Adding to /etc/environment fucked the system
Continuing tests
Well modifying /etc/environment properly seems to be the path, but now the listener doesn’t start…
Seems easiest way is just to declare variables under LCA (/etc/init.d/gctilca) script… as bublepaw suggested
you can also add variables to run.sh in DBServer directory and change application option ‘command line’ in CME from multiserver to run.sh. This should also work just fine and doesn’t need to restart lca if this is case for you.
Gee Kubig how continuously you change Database client software? You add only to LCA script the Oracle variables needed and same to bashrc. So how many? 2.
Why would adding another global environment variable would affect the already existing ones? Sometimes you really just exaggerating everything man… chill down a little bit… and coming from me says a lot lol…
Don’t try to find the 5th leg to everything you see because it is not like that…