Hi Bela, > > I would like to use a distributed map to share state among mbean > > services running on different nodes in my cluster. First I thought > > about using org.javagroups.blocks.DistributedHashtable > > but that would require me to open a new Jchannel which I > would rather > > avoid. So I looked at the code of > > org.jboss.chap8.ex3a.service.DistributedCacheService and > > org.jboss.ha.framework.server.util.DistributedTimedCachePolicy and > > realized that I could just use the distributed map service > that jboss' partition implementation provides.
> Alternatively you could use the new cache, which will replace > that thing at some time in the future. It uses itsown channel for now, > but we will make it use HAPartitionImpl (it's on the tod list). Where is this code for the new cache? Into which jboss version will it get integrated 3.2.x or only the new 4.x? > > Now my last concern would be concurrency. Is the distributed map > > implementation of JBoss or javagroups synchronous over the whole > > cluster? > No, asynchronous. In JBoss, you can choose (in > HAPartitionImpl) between async or sync method calls. If I switch in HAPartitionImpl to use sync calls will this not change the overall cluster communication? If so is this save to do? BTW: maybe more javagroup and jboss related: If I make a synchronized jg call, does that mean it is guaranteed to be executed on all group members before it returns? Regards, Sebastian ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ JBoss-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-user
