We have a custom Deployment of Pentaho using PostgreSQL and Tomcat Apache running within the current version of our proprietary Medical Imaging software (RamSoft). The integration works well, but we have spent months struggling to identify the cause of a major conflict between the PostgreSQL/Tom integration and group policy client. Whenever the PostgreSQL or Tomcat Apache services are running, we begin to see 1 hour + reboot times and gpupdate failures due to the group policy client just hanging for long periods of time with no explanation. Simply disabling PostgreSQL/Tomcat resolves the issue, and allows gpclient to do what it needs to do.

We have enabled all known debugging in Group Policy, PostgreSQL, Pentaho, and Tomcat, performed xBootMgr traces, performed Process Monitor analysis, and Packet Captures,Tomcat_BA Errors.txtTomcat_BA Errors.txt but have been unable to determine the cause of the conflict. We are also working with Microsoft, Apache, and PostgreSQL independently to try and flush out the culprit. After spending weeks analyzing and reviewing our development team's internal notes, I have become fairly confident that the root cause of this problem is related to the way that we deployed Tomcat, and the way that Tomcat/PostgreSQL communicate with each other, but I have not found solid proof that actually indicates this yet.

I have learned a lot about how PostgreSQL/Tomcat are functioning in this environment over the last week, but I am not part of the team that deployed this, and am certainly not an expert on Pentaho, PostgreSQL, or Tomcat. I have been collecting a list of debug error/warnings from the Tomcat logs over the last few days (attached), and I am hoping someone who is an expert on this stuff can possibly review this list of errors, provide an explanation/priority for each, and answer the following questions:

1. Are there any known conflicts with the PostgreSQL/Tomcat integration and GroupPolicy in Windows domain environments? Required Configurations? Workarounds?
2. Does the 'standard' Pentaho Deployment utilize two instances of Tomcat6.exe (one for the DI Service, and one for the BA Service?), or is this a deployment customization that was done by our development team?
3. Are there any special debugging options or monitoring tools that we could use to get more information about what PostgreSQL/Tomcat are doing during the time periods that Group Policy Client is hung?
4. Do you have any suggestions or options that we can try to see if our behavior changes?

Please let me know if there is any additional information I can provide to help.