Which release of MarkLogic is this? On what OS? What's the configured size of the expanded-tree cache?
How much RAM does each host have? So you have 5x10^6 documents on 5 hosts, or 10^6 per host? How many forests per host? How large are the forests on disk? You can let reindexing proceed in the other forests if you like. That won't hurt anything, though they might run into the same error independently. And sooner or later you'll have to finish reindexing the forest that failed. -- Mike On 5 Jun 2014, at 09:34 , kshiiz sarma <[email protected]> wrote: > Hi, > > I am working on clustered environment with 5 hosts. > I am having around 50 lacks documents in DB. > I performed re indexing but after some hours I am getting warning saying > Expanded tree cache is full for one of the host. > XDMP-FORESTERR: Error in reindex of forest .....: > XDMP-EXPNTREECACHEFULL: Expanded tree cache full on host > > Now I have following queries: > 1. 1. Do I need to stop reinexing immediately since I got error for one > host, but I can still see re indexing is in progress for rest of the host? > 2 2. Do I need to wait for reindexing to be complete? > 3. 3. Should I increase expanded tree cache for the D node and then again > start reindexing from the scratch. > > Please suggest. > > Thanks > Ksitiz > > _______________________________________________ > General mailing list > [email protected] > http://developer.marklogic.com/mailman/listinfo/general _______________________________________________ General mailing list [email protected] http://developer.marklogic.com/mailman/listinfo/general
