Sorry I meant clustering of JVM/AppServers, not databases. Can you or somebody else comment on that? - Maybe my question is actually stupid, and all works well as is... Each JVM node will get a separate "block" of IDs from e.g. OPENJPA_SEQUENCE_TABLE, and OpenJPA handles some form of locking/retry on that table to avoid hypothetical collision? All is well, out of the box? Hm... would love somebody to confirm (or deny) this...
Thanks a lot. -----Original Message----- From: Pinaki Poddar [mailto:[EMAIL PROTECTED] Sent: jeudi, 10. juillet 2008 04:24 To: [email protected] Subject: Re: Clustering & ID TableGenerator (OPENJPA_SEQUENCE_TABLE) Are you referring to clustering of databases or clustering of JVM/AppServers? If it is the former, openjpa-slice module that let a persistent unit be connected to multiple databases in the same transaction may be of some relevance. In slice, we designate one database in the cluster as 'master' for the purpose of ID generation. Michael Vorburger-4 wrote: > > Hello, > > How do you do Clustering and ID TableGenerator > (e.g.OPENJPA_SEQUENCE_TABLE) ? > > Will this somehow magically work (how?), or will different nodes "step > on each other's feet" because they'll use the same PK? Can they each > use different "slices" ? > > If impossible, do you HAVE to use Identify columns or Sequence > generators in the DB? I thought that was not such a good idea (if you > have a choice), performance wise, because it requires an extra > round-trip to the DB to get the PK e.g. to set it as FK. Also > https://issues.apache.org/jira/browse/OPENJPA-464 states that > Statement Batching won't work with GeneratedType=IDENTITY - so this > option doesn't seem terribly tempting in order to get ID and > clustering happily married together. > > How do people out there do this with OpenJPA? > > Thank you, > Michael > > _____________________________ > Michael Vorburger, Odyssey Financial Technologies > > > ____________________________________________________________ > > This email and any files transmitted with it are CONFIDENTIAL and > intended > solely for the use of the individual or entity to which they are > addressed. > Any unauthorized copying, disclosure, or distribution of the > material within > this email is strictly forbidden. > Any views or opinions presented within this e-mail are solely those > of the > author and do not necessarily represent those of Odyssey Financial > Technologies SA unless otherwise specifically stated. > An electronic message is not binding on its sender. Any message > referring to > a binding engagement must be confirmed in writing and duly signed. > If you have received this email in error, please notify the sender > immediately > and delete the original. > -- View this message in context: http://n2.nabble.com/Clustering---ID-TableGenerator-%28OPENJPA_SEQUENCE_TABLE%29-tp471147p471566.html Sent from the OpenJPA Users mailing list archive at Nabble.com. ____________________________________________________________ This email and any files transmitted with it are CONFIDENTIAL and intended solely for the use of the individual or entity to which they are addressed. Any unauthorized copying, disclosure, or distribution of the material within this email is strictly forbidden. Any views or opinions presented within this e-mail are solely those of the author and do not necessarily represent those of Odyssey Financial Technologies SA unless otherwise specifically stated. An electronic message is not binding on its sender. Any message referring to a binding engagement must be confirmed in writing and duly signed. If you have received this email in error, please notify the sender immediately and delete the original.
