Brian Elliott Finley wrote:
Andrea,
Was there a problem you were experiencing that prompted this change?
Strictly speaking, it should not be necessary to source those variables
prior to sourcing the functions. We shouldn't be executing anything at
the time that we source the functions -- they all get called at a later
point. And by the time they're called, the variables will have been
sourced.
The variables.txt sourcing is needed by the send_monitor_msg function...
Otherwise switching from the rcS script to the autoinstall script causes
the lost of the previous variables, since clients informations are
evaluated only the first time and they rest declared in env variables...
I don't see any real problem with moving the sourcing of the variables
up to the new location other than the fact that "shellout" is a function
that is not available until the functions are sourced. ;-)
Very true!! the shellout function is not declared at this point... we
could perform the sourcing without considering the shellout, so simply:
. /tmp/variables.txt
This is what we are doing also for functions, I don't see any problem
doing so... what do you think?
Regards,
-Andrea
If we were experiencing a problem, and this fixes it, great -- let's
just do away with the "shellout" bit on the variables line. If there
was no problem, I'd say let's go ahead and leave the variables sourcing
until after the functions sourcing.
Cheers, -Brian
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Sisuite-devel mailing list
Sisuite-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sisuite-devel
