3

My application uses stand alone version of jackrabbit and we wanted to move to embedded mode so that we can cluster it.

I read the requirements on the jackrabbit clustering site but still confused. Should I be having different home directories for each cluster node. i.e. If I need to configure two nodes, do I need to have ~/node1/repository.xml and ~/node2/repository.xml? Or they can share same ~/node/repository.xml?

Has QUIT--Anony-Mousse
  • 76,138
  • 12
  • 138
  • 194
Manisha
  • 31
  • 1
  • My confusion is the same :) Did you managed to find a nice step by step guide / tutorial for configuring jackrabbit clustering? – AlfaTeK Oct 16 '13 at 18:05
  • Hi, I asked part of your questions again in http://stackoverflow.com/questions/31183179/clustering-with-jackrabbit but I already have a working cluster - so maybe one of my posted links or configuration can help you – cloudnaut Jul 05 '15 at 10:44

1 Answers1

3

As described in the Clustering Overview, "each cluster node needs its own (private) repository directory, including repository.xml file, workspace FileSystem and Search index."

Thomas Mueller
  • 48,905
  • 14
  • 116
  • 132
  • And what happens if FileSystem points to shared database? Can you take a look at http://stackoverflow.com/questions/32076020/clustering-pentaho-c-e-5-x-with-jackrabbit-repository? – Boris Treukhov Aug 18 '15 at 15:30
  • If the FileSystem points to shared database, then you probably get a problem. – Thomas Mueller Aug 19 '15 at 06:30