CSIDE errors after VM hard disk filling up. When saving OSpace the error says "database is locked"

I first encountered the problem making a quick change to a function in an unlocked custom module in the Module Explorer before I realized the VM's disk was full. I got :

13:21:25,827 ERROR Failed to save ngppo to C:\OPENTEXT\module\ngppo_16_0_0\ospace\ngppo.oll: database is locked

and it zeroed out the .oll file.

After freeing up disk space and rebooting the VM I still get the error, and have not found anything useful searching the KB, nor searching for the error message in either the .properties files or within CSIDE.

I tried importing the code to source and saving the change there but the Java VM died and put a massive .dmp file on the disk.

Find more posts tagged with

Comments

Greg Griffiths

sounds like the module is corrupted, do you have a backup of the module / ospace ? CSIDE does keep a backup itself as well, so you may be able to use that if you don't have another source.

David Templeton

Please open a support ticket so more detailed information can be retrieved. The real question is are you trying to write directly to the oll, or change the source code and having it recompile?

David

Mayank Gupta

Did we get solution for this? We are aslo facing similar issue.

Regards,

Mayank Gupta

Dave Carpeneto

Hi Mayank - there was an issue with this in CS versions 16.2.5 - 16.2.7. In another forum post from you yesterday it looks like you've recently moved to Content Server 16.2.7. Is that correct? If so: the recommendation is to move up to the 16.2.8 / March release.

In these prior cases the OLL file also got wiped out (it became a 0-byte file). Did that also happen in your case? I ask because there is another "database is locked" issue that is currently in the process of getting fixed (these will be out in 16.2.9). In that case the error was seen whilst patches were getting applied, and the server start-up failed to complete (restarts do work however). Also: the OLL does not corrupt in this specific error case.

If your situation doesn't 100% line up with either of these sccenarios: please open a support ticket so that we can investigate.

All the best, Dave

Mayank Gupta

Hi Dave,

Yes we are on 16.2.7 and yes we also sometime see Database is locked in the Eclipse Console and after one or two minutes we get the EXCEPTION_UNCAUGHT_CXX_EXCEPTION. Oll file size gets reduced to 0 kb in all these cases.

Do you think if we move to 16.2.8 we might also be able to fix this exception error also ?

Regards,

Mayank Gupta

Hany Samuel

Quoted Mayank Gupta on 05/03/2019 09:56 AM:

Hi Dave,

Do you think if we move to 16.2.8 we might also be able to fix this exception error also ?

Regards,

Mayank Gupta

This has nothing to do with CSIDE version you use, it is a problem introduced in CS 16.2.5 till 16.2.7. You seem to work on the oll directly without importing the source to your workspace which triggers this problem. This is fixed on content server 16.2.8 so if you update this will be resolved. Another workaround if you want to keep 16.2.7 for now is to import the source of the oll to your workspace (disable build automatically from eclipse ) and then restart the VM before working on the oll (please note restart not unlock). when restarted, while the source in your workspace, you will find the oll already unlocked and you can work with it as expected. Give a try to either of the options (update to CS 16.2.8 or the workaround) and let us know how it goes.