Cannot start confluence: JDBC exception on Hibernate data access

[doug]

After updating my Parallels VM software, the VM running CentOS with Confluence is unable to start Confluence anymore with this error page:

Description:

Unable to start up Confluence. Fatal error during startup sequence: confluence.lifecycle.core:mananagedjobs (Start and stop the Managed Scheduled Jobs)

Exception:

JDBC exception on Hibernate data access: SQLException for SQL [n/a]; SQL state [XX001]; error code [0]; could not extract ResultSet; nested exception is org.hibernate.exception.GenericJDBCException: could not extract ResultSet

From reading another poster with the same issue, I get the feeling my database got corrupted by the VM software update (perhaps the VM did not shut down properly).  I don't have backups enabled in Confluence.  I am using Postgres (not H2).  Is there an easy remedy?

[Diego]

Hello there!

As I understand, something went wrong during VM upgrade and now Confluence and Database are not able to communicate, correct? Before performing any kind of change to the environment, it is always a good practice to back up everything. So if disaster strikes, we are ready!

Doug, this does indeed looks like a big problem with your database just like you pointed out.

There are some things we can check though. Here you go:

1. Database version
2. Confluence JDBC driver
3. Confluence confluence-cfg.xml file (if it is pointing to the correct database
4. If the database service / server are starting correctly

Also, could you clarify a few points for us?

1. Was Confluence running when this VM upgrade happened?
2. Which version of the database were you running?
3. Which version of the database are you running now?
4. Which version of Confluence is currently running?
5. Is there any other application that connects to a database on the server that got updated?
6. Is the updated VM still reachable?
7. Has the VM address / name changed after this upgrade?

Looking forward to your reply Doug!

[doug]

Hi Diego,

Sorry for taking so long to reply. Have been on business trips.

Unfortunately, Confluence was running when the VM was upgraded and I suspect that is the trigger to the file system corruption.  I once had a MacOS VM get completely trashed by a VM update previously.  

So, just a warning to the community: be very careful when using Parallels for VMs running server apps like Confluence!  Their update process is certainly flawed when they automatically hibernate the VMs.  That process can get into an infinite loop and never end (I waited 8 hours on the last update).  When this happens, you most assuredly have data corrupted somewhere.  I'll probably change to VMWare soon.  Best advise: manually shutdown all VMs and back them up before letting the VM software update itself.

I was able to find a backup of my VM image that was only 1 month old.  Changing to that got Confluence running again with only the loss of a handful of Gliffy diagrams.  I plan to stick with this as my recovery rather than try to recover the database.

Thank you for the response!