The GridFS menu also provides a selection to delete GridFS files. Sadly, the response document only indicates whether the delete command executed -- not whether the deletion actually occurred. You have to watch the document count that is displayed next to the fs.files collection in the database navigation tree displayed in the left-hand pane.
Tools. This menu also differs depending on whether a server, database, or collection is selected. At the server level, you can enable or disable auto-balancing (the process that MongoDB runs to keep shards balanced across the members of a sharded cluster). Or you can regenerate the configuration database, which is the server's internal description of the cluster configuration.
The Analysis display is a quick way to reveal trouble spots or slack resources based on performance characteristics. We can see the analysis being performed, and the resulting heat map and detail below.
Finally, at the collection level, the Tools menu has entries that let you validate the structure of the collection, rebuild a collection's indexes, compact a collection, set various database options associated with collections, and so on.
A failure in UMongo can sometimes result in a Java exception being thrown. For example, if you try to upload into the GridFS a file that doesn't exist, you don't get a nice, neat error message -- you get a stack dump. The system displays a dialog box showing not only the details of the command that had been issued, but a Java stack trace of the exception as well from which it's pretty easy to deduce what went wrong. Although this is disconcerting -- when it first occurred, I wondered if something had broken -- it doesn't derail UMongo. Just close the dialog box and continue working.
While this article was being written, a new version (1.4.0) of UMongo was released. Version 1.4.0 adds support for MongoDB's user roles by providing a graphical interface that mimics the capabilities to be found in the mongo shell's addUser command. It also adds controls for the sharding balancer. For example, version 1.4.0 lets you set the time window when the balancer runs, so you can schedule the balancer to execute during low-traffic periods of the day. The new version also has support for TTL (time to live) collections. When a document is inserted into a TTL collection, the system appends an insertion time. After the specified TTL passes, MongoDB will delete the document.
Sign up for MIS Asia eNewsletters.