Guillermo Averboch wrote:
Knut,
Actually, it does not work. If you set derby.storage.pageCacheSize after
loading a database, the setting affects that database too. In the end, the last
setting affects all databases.
Hello Guillermo,
I tried the approach suggested by Knut Anders, and it seemed to work for
me when I tested it on trunk. I added a trace print in the cache, and
could see that it got initialized with different values. As the maximum
size is kept in a final variable, the only way it can change is if the
database is shutdown and rebooted. Now, this is only valid for the new
cache implementation, but I would believe the approach should work for
the old one too.
So, a few question that might help to figure out what is going on:
a) Are you sure the databases are not rebooted after the initial boot?
b) Are you sure you set the system property in the VM process that
boots the databases?
c) Which driver are you using?
I assume you have written a Java class that boots the databases for you
and possibly starts the network server?
regards,
--
Kristian
I learned this the hard way, after everybody started complaining about how
awfully slow the system was...
Guillermo
