We're potentially in the market for some Jackrabbit consulting help. We're currently architecting a site with high-performance/high-availability requirements. In a previous project, we implemented a CMS based on Jackrabbit using an Oracle repository operating in clustered mode (version 2.2.1). Although it works for the most part, we do see occasional issues in two areas: 1. A cluster node will stop processing revisions. 2. Since our administrative application runs on multiple servers for HA, we have seen race conditions where cluster nodes may get nodes with different state - assume due to normal cluster sync latency and the fact that there's no way to get a cluster-wide lock due to JCR-1173.
Although #1 may be fixed in later versions, it appears to be fundamentally problematic to try to run CMS applications in separate cluster nodes. We're interested in getting some help with architecting a solution that: * Eliminates these problems. * Provides for high-availability for applications doing both read/write (admin) and read-only access to the repo. Pointers welcome - thanks. Tom Palmer Lead Principal Technical Architect AT&T Hosting & Application Services | 2000 Perimeter Park Drive, Suite 140 | Morrisville, NC 27560 Office: +1 (919) 388-5937 | Mobile: +1 (919) 627-5431 [email protected]<mailto:[email protected]> Confidentiality Notice and Disclaimer: This e-mail transmission may contain confidential and/or proprietary information of AT&T that is intended only for the individual or entity named in the e-mail address. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or reliance upon the contents of this e-mail is strictly prohibited. If you have received this e-mail transmission in error, please reply to the sender, so that AT&T can arrange for proper delivery, and then please delete the message from your inbox. Thank you.
