Enterprise Vault File System Archiving has over the last few years had many improvements made to it to really just the ‘Enterprise’ title. One of the best of these is a checkpoint mechanism. In this post I’ll explain a little about how it works and how it can really help.
If you have a large file server with 10,000’s of folders, and millions of files, it really doesn’t matter how big you make the archiving window it is unlikely that the archiving task will manage to get through evaluating every single file. Prior to the introduction of the checkpoint mechanism this was a big problem for Enterprise Vault administrators. What would happen is that each time the archiving task started it would scan through folders and files and start archiving and processing them.. but it would never get to some folders.
People tried many things to get the task to complete a full pass, such as making the window bigger at weekends, or by splitting volumes and targeting in different places. But these didn’t always work.
With the checkpointing mechanism what happens is that every minute or so information relating to where the task is currently processing, including the actual file, is written to an .XML file on the EV server. When the task stops, the file is also updated. This means that the next time that the task starts, it can read this XML file and work out where it should ‘resume’. After a little bit of processing, that’s exactly what the task can now do.
Here is an example of the file:
It should be noted that if you run the archiving task manually from the ‘targets’ section of the VAC you might lose the checkpointing information. You get the following popup:
After the pop-up you get the normal ‘run now’ dialog asking if you want to run in normal, or report mode.
Lastly you can configure whether or not checkpointing will function on the properties of the FSA task as shown below:
Image credit: Mark Cartwright