In addition to supplying individual SQL files, have a single file that will execute all SQL (in the correct order) for setting up a new database so that file can be submitted to DBA for execution.
Allow setting up of new database via Neuron Explorer, even if the database already exists. Our DBA will not allow us to create databases, but we can update / make changes to a database that is already created.
A more consistent UI for Neuron Explorer
Work Online / Work offline is confusing. There are several features that overlap. These should be distinct or at least clearly noted.
Active deployment group should be determinable from any screen in the UI
Current active deployment group should be selected (and default value) from a single place. Anything in the Activity group should default to the selection of the server in the toolbar.
Database reports should be referenced via deployment group name and not database server. With multiple deployments, it is an unnecessary requirement to have to remember which database server goes with which deployment
ESB Diagram is often inaccurate... either correct or remove
Make Message History report more selectable / sortable.
Make Active Sessions report more selectable / sortable (Currently sorting only works for a single page and not for the entire result set)
I think that I understand the intent of being able to look at other deployment group instances in the activity section, but it often obfuscates the UI. Server in toolbar says one thing, Yellow info bar says another, and anything in Activity section could be neither of the above. Maybe ESB Monitor / Dashboard is an application unto itself?
Store messages in the database in XML (not binary) form. There is valuable information within the message that is difficult to extract without writing an application to deserialize the message data. All data should be queryable via SQL. I view with the message data in XML would satisfy me if it must be stored in binary format for performance reasons
More documentation on API. The 2.5 releases are going in the right direction, but still fall way short of expectations. See MEssageSequenceEventArgs, PipelineEventArgs, *EventArgs for examples.
Add search and index to help files
Neuron does not validate sub topic values. IMHO, it should
In pipelines, allow the publish step to also set source value in addition to topic value
Provide built in durable on-ramp to Neuron. This would be a method to send messages to the ESB, even if the ESB is unavailable (ala MSMQ)
Document database schema / procedures / functions / etc.
Adapters configured for publish only, should not be considered as subscribers (receivers) of messages
Unify endpoint configuration... services are nothing more than "special" adapters.
Add / document configuration API so that repetitive manual entry can be automated (at least to some degree)
Allow for "more natural" pipeline ordering, instead of having to delete and then add in required order
Do not "auto start" Activity Real-Time / Running History graphs. If the defaut deployment group is not available, the application hangs and occasionally becomes unreliable.
Add a visual queue that Neuron has picked up changes to configuration and is now using current values.
As extra nice would be to list what changes have been picked up. At the very least a notification of what changes require a restart.
We have set up an ODBC adapter to poll a stored proc.We found that if the stored proc has a temporary table defined the rows returned are always 0.Any idea why this would be and what we can do to get around it?See More