Grouper Users - Open Discussion List

[Chris Hyzer <>]

Another thing... its on our roadmap to improve availability of Grouper (post 
2.0).  I think it is on the table to be able to configure multiple database 
connect strings, and/or retrieve nodes from dns-srv or something, and have 
Grouper's data layer have a try/catch with timeout (could configure to only 
be readonly) so it could not have to rely on database load balancers or 
drivers etc...  my understanding with Oracle ONS is if there is a connection 
checked out of the pool, and a failure of a database node happens on that 
node, then the connection will not recover, so even with Oracle I think this 
type of error handling could help

Thanks,
chris

-----Original Message-----
From: 

 
[mailto:]
 On Behalf Of Peter Schober
Sent: Thursday, June 09, 2011 3:11 PM
To: 

Subject: Re: [grouper-users] Grouper and database failover

* Peter DiCamillo 
<>
 [2011-06-09 20:33]:
> Thanks for the information. Sorry for the premature request-- it
> turns out that our DBAs don't have RAC installed, but use Oracle
> Data Guard. So I'll be investigating that, and would be interested
> in any information regarding use of it with Grouper.

(Not Grouper, but...)
We're using our Shibboleth IdP with a DB in a Data Guard setup (via a
container managed connection pool, not by the application) and it
works well. All that's necessary is to supply the complete
tnsnames.ora for the service in the connection string, very similar to
what Chris just posted:

url="jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=db.example.org)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=otherdb.example.org)(PORT=1521))(LOAD_BALANCE=
 off)(FAILOVER = on))(CONNECT_DATA=(SERVICE_NAME=FOO)))"

There's a paper availabe from Oracle, IIRC, for what database clients
can do, depending on the technology, to enable switchover and failover.
-peter