Hey guys, I've made a bit of progress on the win32 regression scripts.
In order to use the win32 bacula regression scripts, it is important to have
-some unix tools (mainly sed). To make things simple, I downloaded UnxUtils
-from http://sourceforge.net/projects/unxutils
+some unix tools (such as sed, grep, and diff). To make things simple, I
+downloaded UnxUtils from http://sourceforge.net/projects/unxutils
-Extract this somewhere on your hdd ( I extracted to C:\unxutils ) and add
-the usr\local\wbin path from wherever you extracted the utils ( for me it
-was C:\unxutils\usr\local\wbin ) to your {$PATH} on your windows box. If you
-also want to make use of the shell (from unxutils) you may want to add the
-./bin directory as well (or just copy the sh.exe file to usr\local\wbin).
+Extract this somewhere on your hdd and add the the files in usr\local\wbin to
+c:\regress\tools and then add c:\regress\tools to your {$PATH} on your windows
+box. If you also want to make use of the shell (from unxutils) you may want
+to add the ./bin directory as well (or just copy the sh.exe file to
+usr\local\wbin).
Now that you have a working toolset (with sed) we can continue setting up
regression tools. Compile the Ming32 version of bacula (from linux) and then
binaries from the bacula sources and create a local bacula testing platform.
-----------
-At this point all three of the bacula daemons fail miserably mentioning
-something about bacula.dll. I will try figuring out what else is wrong
-because when I installed the same build of bacula with the installer the
-system ran fine. I'm guessing there are some outdated test configurations
-and filenames causing bacula to go on the fritz.
+After the "make setup" all the configuration files are blank, so you must
+fix that before running the scripts.
+
+With the above, most of the regressions scripts seem to function flawlessly.
+
+[KES] The ones that fail have probably been modified on the Unix side
+ after the port was done, so they may need some minor updates.