There are a couple of things to investigate here - the log parts you've sent indicate that when taking the backup, there were no full backups to append an incremental to, so the backup process is taking a full. This might be due to retention rules being applied, but if they were, it would be listed in the log or previous log. We only delete files for retention reasons when backups happen, so it will always be noted in the log.
For the IP address, Agents backup directly to the repository, so the Site Manager server sends down a list of the IP addresses (obtained via DNS) to the Agent so that if there are any issues with accessing the repository by name, the Agent can fall back to IP address. This will be noted at the top of the log though.
Can you confirm that the IP address is the correct one for the repository and that all the full backups are present on the path being backed up to?
Another possibility is that the disk layout has changed since the last backup - if this happens, we will start a new backup set as the full an incremental is based on has to contain the same partitions as the incremental. Depending on your backup definition, this may include removable disks (there is an 'Exclude removable disks' rule that can be used to exclude these) - it might be worth checking the disk list displayed in site manager to see if there are any unexpected changes in layout.
See our reviews on