Build 25 was built based on Monday midnight workspace snapshot so this will be different from your current workspace. Other than that, there are differences in build process when it comes to assembling IPS enabled vs. non-IPS enabled distributions but this should not matter unless something really bizarre happened...
Thanks,
Snjezana
----- Original Message -----
From: bobby.bissett_at_oracle.com
To: dev_at_glassfish.dev.java.net
Sent: Wednesday, October 20, 2010 12:16:03 PM GMT -08:00 US/Canada Pacific
Subject: P1: cluster upgrade broken in b25
Hi all,
These steps work fine from my workspace (updated this morning), but fail with build b25:
http://wikis.sun.com/display/GlassFish/V3.1ClusterUpgradeExample
Can someone tell me if the promoted build is created any differently than the normal build I do locally? Or tell me what revision was used to create b25? I want to see if these issues are actually fixed now or if they're systemic to some build process that I don't know about.
For instance, I haven't seen this DAS error in a long long time:
SEVERE: class com.sun.hk2.component.LazyInhabitant upgrading domain.xml failed java.lang.RuntimeException: org.jvnet.hk2.config.TransactionFailure: Constraints for this bean violated.
Message = gmsMulticastAddress may not be nullgmsMulticastPort may not be null
More importantly, the applications that were deployed to the cluster have disappeared after the upgrade. They're not there in the admin gui and I get 404s when trying to hit them directly.
This would normally be a blocking P1 issue for me, but I can't reproduce it in my workspace. Art has also seen this with b25 and not with his workspace. Since QA testing acts on promoted builds, cluster upgrade testing is blocked until I can figure this out.
Thanks,
Bobby
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
For additional commands, e-mail: dev-help_at_glassfish.dev.java.net