Thanks a for quick reply. Alexander Klimetschek wrote: > > On Thu, Jun 19, 2008 at 11:34 AM, Roman Puchkovskiy > <[EMAIL PROTECTED]> wrote: >> Is it correct that all corresponding Persistence Manager's configs should >> be >> the same? I.e. both versioning Persistence Managers should store their >> data >> in the same database tables (e.g. with prefix versions_), and so on? > > Yes, you need a database that handles this concurrent access properly > (which should be the case with most of them - not sure about the > correct isolation levels needed). > >> Is same correct for File Systems? Should, for example, >> Repository/FileSystem >> on A have same config as Repository/FileSystem on B? > > Jackrabbit does not have a file system based persistence manager that > allows clustering (yet). >
I'm sorry for ambiguous question. When asking about FileSystems, I meant FileSystem elements in the repository.xml (i.e. objects which implement org.apache.jackrabbit.core.fs.FileSystem interface), not file system as a generic OS understands it. Cluster must have a shared Journal, it's clear. But i'm not sure did I understand the cluster configuration guide correctly or not, because requirement to store PMs data of different cluster nodes together looks a little strange. Alexander Klimetschek wrote: > > > Regards, > Alex > > -- > Alexander Klimetschek > [EMAIL PROTECTED] > > -- View this message in context: http://www.nabble.com/PersistenceManager-and-FileSystem-config-for-cluster-tp18000628p18005739.html Sent from the Jackrabbit - Users mailing list archive at Nabble.com.
