Hi Brian,
indeed, from scripting point of view, openser 1.0.0 is still reasonable
compatible ser 0.9.x. You may find that some things got easier to be
done due the pseudo-variable support and their interaction with AVPs.
Regarding DB transition - I would say only the format of the
location/aliases table (used by usrloc) were affected - new column
"socket" was added (a short look on the table format in
scripts/mysqldb.sh will be useful); also do not forget to update the
version of these tables!
So, adding a new column and changing their versions will do it.
for more about the differences, see
http://openser.org/diffs-0.9.0.php
http://www.openser.org/release-1.0.0.php
regards,
bogdan
Brian McCrary wrote:
Hello,
I am currently using SER 0.9.4 in a production environment. I've
patched it with various modules that OpenSER has, but would like to
convert over to the OpenSER project to be able to use a lot of the new
functionality. I've comiled the latest CVS source (the main release has
Solaris compile problems that have been fixed in CVS, hence the CVS
source).
The process doesn't look too difficult, a few things to be changed in
the config file, namely break and return. However, I'm not sure what's
the best way to convert the MySQL tables over.
My tables now are in the ser database instead of the openser. I
imagine I could duplicate the ser database into one called openser, and
perhaps run the openser_mysql reinstall script to upgrade.
I've looked and can't really find any documentation on converting over,
so if anyone that has done this conversion before has any ideas or what
to expect I'd appreciate it. My goal would be to have openser run on a
test server, but be able to access my production MySQL server, but using
a unique database name (openser instead of ser) so I can iron out the
config file, and everything else.
Thanks,
Brian
_______________________________________________
Users mailing list
Users(a)openser.org
http://openser.org/cgi-bin/mailman/listinfo/users