We are dealing with an issue where our Jackrabbit server attempts to start before MS SQL Server is ready. The login fails, Sling is unable to instantiate the persistence manager, and Jackrabbit is left in an unusable state. We are working on the timing issue but we'd also like to harden Sling against temporary outages of the database. I've looked throug the documentation and some of the code chain leading to this exception but I don't see any attempts at retrying this operation; it appears to be one shot, hit or miss. Is there any way to get it to retry under these circumstances? I was able to do a live workaround by stop/starting the bundle 'Apache Sling Jackrabbit Embedded Repository'; this refreshed the persistence manager and made the repository available, but it's a rather rough solution. Thanks!
