Daniel,
I agree with this proposal.
In addition to the review of the columns that need to be kept, I have
three additional suggestions:
1. The names of the columns should be made consistent.
For example: column name associated with the last_updated time of a record.
2. Additional columns should be added to existing tables where the are
needed to enhance OpenSER
For example: addition of a "description" column to the "trusted" table.
3. New table(s) might be added to enhance OpenSER
For example: a table listing valid "grp" entries. This function is hard
coded into openserctl today. An extension of this idea would be to
control the addition of AVP attributes in a similar fashion. It does
not seem logical that openserctl "controls" grp entries but not AVP
attributes. Alternatively, grp checking from within openserctl could be
removed completely (I would be opposed to this).
Regards,
Norm
Daniel-Constantin Mierla wrote:
Hello,
at a moment it was in discussion whether to change the database
scripts to create only the tables needed by openser and the rest using
a special parameter, something like:
- opensermysql.sh create openserdb (same as "opensermysql.sh create")
- creates only the tables needed by openser
- opensermysql.sh create serwebdb - add the tables needed by serweb
and adjacent applications
- opensermysql.sh create alldb - create all the tables
Is this a viable solution, or should we stick to current version? Many
of us have their own admin/provisioning interface and the extra tables
are useless. Let's see what would be more convenient and take a decision.
Also, some columns must be reviewed, to see if they are still required.
Cheers,
Daniel
_______________________________________________
Devel mailing list
[email protected]
http://openser.org/cgi-bin/mailman/listinfo/devel
_______________________________________________
Devel mailing list
[email protected]
http://openser.org/cgi-bin/mailman/listinfo/devel
