This one time, at band camp, Aditya Akella said:
AA>Hi Steave,
AA>
AA>Here is a link. Check the "TIP:" in that after the 4th point.
AA>
AA>http://www.mail-archive.com/[EMAIL PROTECTED]/msg04437.html
AA>
AA>I don't know how to fix the problem but this might help you and if
AA>you find a wayout please post it on the mailing list.
AA>
AA>Hope the above link helps.
AA>
AA>-- Aditya
AA>
AA>-----Original Message-----
AA>From: Steve Chernyak [mailto:[EMAIL PROTECTED]
AA>Sent: Friday, June 06, 2003 12:10 PM
AA>To: [EMAIL PROTECTED]
AA>Subject: [castor-dev] Missing records from OQLQuery
AA>
AA>
AA>Im having a problem retrieving records from the database using OQLQuery.
AA>I am running an a clustered WebLogic 8.1 environment using 0.9.4.3 version
AA>of castor.
AA>
AA>The problem:
AA>
AA>Executing the same query from different servers in the cluster results in
AA>different size results being returned. A record is missing on one of the
AA>servers. This problem goes away after the servers are restarted. Also, this
AA>doesnt happen all the time.
Steve,
Aditya's suggestion to look at Alexey's post to the list is a good
one. In the Tip, Alexey is making light of the fact that Castor has
no distributed caching mechanism. This means that if Object A is
changed on AppServer A, there is no automatic distribution of this
change to AppServers B and C. I plan to remedy this in the JDO
refactoring that is currently being planned.
On a side note, Alexey is correct that the capacity of the count-limited
cache type used to be 30. Just within the last month or so I increased
this default to 100.
Bruce
--
perl -e 'print unpack("u30","<0G)[EMAIL PROTECTED]&5R\"F9E<G)E=\$\!F<FEI+F-O;0\`\`");'
-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
[EMAIL PROTECTED] with a subject of:
unsubscribe castor-dev
