Max, > "attempted to deregister a transaction that is not currently registered" is > an error thrown by the datastore api when a call is made to commit a txn and > there is no txn to commit. Is it possible you have some path through your > code that would result in calling commit() more than once or calling > commit() after a rollback()? On the first occurrence of this exception, the request triggered application to be loaded and the request processing time was getting to 30 seconds (according to the log total time for the request was 30.417s). Is it possible that appengine's rollback logic for the 30 second limit started kicking in just about the time the app tried to call commit?
> I can't explain the partial data you're seeing - transactional writes are > atomic from your app down to the datastore down to the storage layer that > the datastore is built on. Since you had 2 cron jobs executing concurrently > is it possible that the writes interacted with one another in some > previously unforeseen way? It is possible. The request processing logic runs within a spring transaction context. There is a third party library which is using low level api to do updates so maybe these are not behaving in the way that I'm expecting. I will take a more closer look into if the requests are properly being processed within a transaction. Thanks, Len -- You received this message because you are subscribed to the Google Groups "Google App Engine for Java" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-appengine-java?hl=en.
