When configuring a repository, you have to specify retention policy for mailbox items. This section explains how this setting is applied to mailbox data.
Let's assume retention policy for repository was set to 3 years (default value), and you have configured a backup job targeted at this new repository.
This will work as follows:
- When the backup job starts, the program will first analyze the metadata of the mailbox items selected for backup (before copying the item data itself). In particular, it will check the last modification date for the item and calculate item "age".
For example, a job runs at 10:00 AM on September, 15, 2017, and last modification of an e-mail was made at 10:00 AM on September, 1, 2014. This means that e-mail "age" is 3 years and 2 weeks.
- This "age" is compared against the value set in the Retention policy for mailbox items for repository - that is, 3 years in our example. If the "age" exceeds the specified interval, then such mail item will not be backed up to the repository.
In our example, the e-mail item is older than 3 years, so it will not be backed up.
On the contrary, an e-mail that was last modified at 11:55 AM on September 16, 2014, will be successfully archived to the repository.
So, this setting is first applied when the program analyzes source mailbox information before storing mailbox item data to the repository. This approach allows you to decrease the amount of data transferred from source to target.
Next, this setting (3 years or other) is applied to the backup data stored in the target repository, as follows:
- When all mail items "young" enough for backup are stored to the target repository, the program checks the retention policy application schedule set for this repository.
This setting defines how frequently Veeam should analyze the "age" of mail data in the backup, comparing it against the specified threshold (retention period), which is 3 years in our example.
- When the policy application process starts, the retention period starting point is set to the current date/time. Obsolete data is removed from the repository automatically, using the same logic as described above:
- As for an item, it will be removed from the repository if its "age" exceeds the threshold, that is, its retention period expired.
- As for a datastore, it will be removed from the repository when the last remaining items in the archive have their retention period expired, and no data arrives from Exchange server by that moment.
Note that this process schedule is independent from the backup job schedule.
Backup to our example, this means that the e-mail which was last modified at 11:55 AM on September 16, 2014 (and, therefore, successfully backed up by the job run at 10:00 AM on September 15, 2017) will be analyzed daily at 12:00 AM. The nearest analysis that starts at 12:00 AM on September 16, 2017, will detect that this e-mail became older than prescribed by retention policy (3 years) and must be removed from the repository.
Retention policy is applied to backup repository, not to backup job. So, if you want data processed by different jobs to be preserved for different periods of time, then you should target these jobs to different repositories, which have appropriately configured retention policies.
For example, a company needs to preserve Customer Service mail for 3 years, and Marketing mail - for 2 years.
Then she targets the CS Mail Backup job at the CS Mail repository, and Marketing Mail Backup job - at the Mrk Mail repository.