I have Confluence 4.1 installed and running with an external Oracle db. I am trying to upgrade and have been running into roadblock after roadblock. I decided to install 4.1 on a seperate server and try to upgrade it to make sure this works before i touch my production version. I tried to make the copy of my production version as suggested in the documentation and that did not work so i figured id just try and install and upgrade an evaluation version. I keep running into these errors when I first startup confluence after i have made the correct adjustments to the server.xml and the confluence-init.properties files along with placing the correct ojdbc5jar in the lib directory. These errors appear on the first startup of confluence:
this is from the atlassian-confluence.log....
2012-08-21 08:10:57,214 ERROR [main] [atlassian.confluence.upgrade.UpgradeLauncherServletContextListener] contextInitialized 1: Cannot update schema
2012-08-21 08:10:57,215 WARN [main] [atlassian.confluence.plugin.PluginFrameworkContextListener] contextInitialized Not starting full plugin system due to upgrade or licensing errors
A workable workaround to this issue is:
However, there is a workaround for your issue, you will need set the system property hibernate.hbm2ddl.skip_creating_missing_indexes to true to avoid automatic index creation before the upgrade.
This workaround is not part of the official documentation afaik but works until Atlassian will fix this bug in the upgrade!
Im going from either 4.2.11 or 4.2.13...ive tried both. I havent tried 4.2.12.
I dont understand why there have to be so many hoops to jump through just for a upgrade. My 4.1.4 keeps running into known memory issues but I swear that is better than these upgrade issues im having. Let me know if you figure anything out and ill do the same. Maybe together we can help someone else in the future with this post.
Normally an upgrade is very smooth. Atlassian has done its best to make it as easy as possible. But sometimes you might run into strange problems that are hard to tackle.
There is a (very old) bug raised on an issue similar to ours. We could work around the problem with this option in confluence.cfg.xml: "However, there is a workaround for your issue, you will need set the system property hibernate.hbm2ddl.skip_creating_missing_indexes to true to avoid automatic index creation before the upgrade."
So far this is working in my simple recreation of my production version...going to try and apply this to my prod version and see what we come up with. Thanks for the tip. I read that doc on manual installation a 1000 times and must have missed that property to add.
I finally got this straightened out and had to do it the manual way...Here is what i did:
*******************BEFORE YOU DO ANYTHING MAKE SURE YOU HAVE A BACKUP OF YOUR DATABASE AND YOUR FILES. IM NOT RESPONSIBLE FOR SUPPORTING THIS WAY OF UPGRADING ITS JUST WHAT WORKED FOR ME.********************
Obtain the tar of the version that you want to upgrade to from Atlassian's download page.
Unzip the tar to a folder other than where your current Confluence is installed.
In my case I had atlassian-confluence-4.1.4 and unzipped my tar to atlassian-confluence-4.2.13
I copied my ojdbc.jar file in my confluencehome/lib directory to the new confluencehome/lib dir...
cp atlassian-confluence-4.1.4/lib/ojdbc5.jar atlassian-confluence-4.2.13/lib/ojdbc5.jar
For the Confluence home files...
I copied the server.xml file to the new home...
cp atlassian-confluence-4.1.4/conf/server.xml atlassian-confluence-4.2.13/conf/server.xml
None of the settings should have changed in here but double check your database connection information to be sure.
In your new confluence home dir find the confluence-init.properties file inside...
In my case this is atlassian-confluence-4.2.13/confluence/WEB-INF/classes/confluence-init.properties
Look to the bottom of the file and find: confluence.home= Uncomment it and set it to your Atlassian home dir. For me this is confluence.home=/opt/atlassian.
For the Atlassian home files...
Go to your Atlassian home directory (data folder) For me this is /opt/atlassian
edit the file confluence.cfg.xml
if it is not already there add this line:
This will prevent Confluence from trying to reindex tables that have already been indexed. You will still see some new indexes created.
Shutdown your "old" version of confluence.
Navigate to your "new" version and start it. For me this is /opt/atlassian-conflunce-4.2.13/bin/start-confluence.sh
Be sure to tail 2 different files at the same time to observe what is happening:
tail -f /opt/atlassian/logs/atlassian-confluence.log <--- This is the "system" log that you see referred to in the Docs from Atlassian. Lots of info in here
tail -f /atlassian-confluence-4.2.13/logs/catalina.out <---log of TomCat...you will see when Confluence is actually started up in here.
Once you see that Confluence has started up you should now be able to log in and see the update has taken place.
Atlassian Summit is an excellent opportunity for in-person support, training, and networking.Learn more
...there's anything I've learnt from working, it's that people are lazy! No offense to anyone reading this, but it's true and we can all admit it. The easier you make something for someone, the more...
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG
You're one step closer to meeting fellow Atlassian users at your local meet up. Learn more about AUGs