i know jboss can't run on J2SE-1.5.0-Beta-1 due to a JMX issue, but how about Beta 2,
just released recently? and if not Beta 2, is jboss working with sun and will the
issue be resolved in the final release of 1.5.0?
would be interesting to get some more information on the issue as well.
see h
if i understand your discussion, i believe i may be having a similar problem which i'm
trying to tackle at the moment, which is basically the question:
"can a CMR field be part of a CMP primary key?"
and if so, how does one do it? i'm also wondering if xdoclet would do it properly, if
it's poss
i think i see the point ---
CMP's must be set in the entity bean's ejbCreate method, which returns the primary
key. and CMR's must be set in the beans ejbPostCreate method, which is called after
creation. thus, the CMR fields formally can *not* be part of the primary key (correct
me if i'm wron
what i mean is: if the CMR field (called 'foo') which is being mapped to the same
database column as a CMP field which is part of the primary key, then doing a setFoo()
on that CMR field would in effect be changing the primary key too, which is a no-no.
thus, i would not expose that setFoo() CMR
this is also related to a question i was about to ask, which is:
"i know there is caching for entity beans based on their primary keys, but is there
also caching based on previous executions of finder methods?"
[correct me if i'm wrong, please]
for instance, if i have an 'account' entity bean a
thanks for the info, and i'm already well aware of the commit options and the dbindex
xml-element. those certainly can improve performance.
but i believe those things just pertain to the database or primary-key caching itself,
right, and not to the caching behaviour for finder methods in particu
hi folks,
i'm running a finder method twice in succession, i.e., one *immediately after* the
other, and the results are different each time --- clearly seems like a bug.
i'm doing something special which may be causing the trouble, despite its being
suggested in this forum: i'm sharing a databa
sorry, i described it slightly incorrectly:
although this entity bean has a CMP primary-key field and CMR field that share a
database column, the CMR that the finder method is navigating is *another* CMR field,
which has its own database column on the 'other' side of the relationship. thus, it
hi there,
i prepared a zip file which is a complete jboss ejb project with a client, intended to
[eventually] exhibit the bug i described in previous posts. the zip file is available
at http://xoba.com/debug-2.zip and basically just contains three entity beans and a
single client class. a 'prop
i updated http://xoba.com/debug-2.zip to reflect the state of that project after its
built, so that deployment descriptors created by xdoclet etc etc are all present now.
this may be more convenient than depending on folks to build it for themselves. thanks
in advance --- mike
http://www.jboss.
glad to do so --- i may delay a bit while i investigate some more, however --- hope
that's alright. thanks - mike
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3824620#3824620";>View
the original post
http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=3824620>Reply
to
hi folks,
is there a way to register for events having to do with jboss' caching of entity beans?
is the mbean InvalidationBridge or InvalidationManager in the jboss.cache domain
somehow involved in that?
couldn't find any info in the $$-docs on the topic.
thanks,
mike
http://www.jboss.org/
hi folks,
i found a bug in how jboss-3.2.3 is translating ejb-ql to sql for the Hypersonic
database. actually, i think its a more general problem since i usually use mysql with
jboss, but was forced to try hypersonic because sql was bombing even more dramatically
with mysql (jboss was basically
thanks a million! i really appreciate it. now i'll go back to my original ejb-ql
finder problem (earlier in this forum) and see what further progress i can make. BTW,
i'm going to make an effort to understand the CMP part of the jboss code so i can be
more helpful and productive if or when i do
one more quick question:
was it fixed yesterday or earlier, or just this morning? because a few minutes ago i
downloaded a fresh jboss-3.2-snapshot from sourceforge, built it, and still saw the
bug; namely, that for both ejb-ql phrases 'IS NULL' and 'IS NOT NULL' on my CMR's,
jboss is executing
i apologize if my answer is not the most authoritative, because i'm not a jboss CMP
developer, but my understanding is that this kind of class can create or is
responsible for creating primary keys for entities. i personally use a standard one
for mysql which helps with auto_increment tables (ge
i rather believe it will be stable for quite some time, since (1) the jboss interface
his class probably implements will not likely change anytime soon, and if it does, it
would most likely remain backwards compatible (potentially requiring a recompile in
that case), and (2) the same reassurance
hi folks,
i think i found a jboss bug (3.2.4RC1) which basically renders my
application dead in the water, and it relates to something pretty
fundamental in ejb-ql CMP/CMR and finder methods. i also posted a
bug report on sourceforge on this.
its basically the next step after my previous topics
rather than declare a return type of java.lang.Object, i think you rather want to
declare the local interface for the bean. this may or may not have anything to do with
your particular bug, however.
-mike
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3824930#3824930";>View
the ori
19 matches
Mail list logo
