Hi, can you please try to find out more details on what slows down the startup. e.g. you could take a thread dump while it is starting, that will should you what's going on.
maybe the copied instance is replaying the whole journal, not sure though... regards marcel On Wed, Apr 8, 2009 at 14:14, Kurz Wolfgang <[email protected]> wrote: > Hello everyone, > > We have a Jackrabbit setup with multiple clusters for development purposes. > At night the persistence managers and datastore are backuped from the live > system to the development system. Just by copying everything over to the > other location. > > Now the problem is that in the morning the startup of the clusters take for > ever. But after they successfully started up for the first time it's just a > matter of seconds to restart them again. > > So my guess is that the indices have to be rebuild at the first startup. But > even copying everything including the indices didn't help speed up starting > the clusters. > > We are getting to a point where the maximum Server timeout interval isn't > enough to get the servers up and running. > > Anyone have an idea what going wrong? Or what we could do to fix this? > > > >
