My server is a limited space mailserver set up to filter spam. Most of it's mail is passed on ot an exchange server so it is not stored locally. I have a handful of pop accounts that are stored locally. For years I've used mbox type mailboxes that I'm familiar with and can access on the server when needed using the old mailx program. They use a variety of clinets, typically Outlook and smartphones.
I'm considering changing to IMAP because I need a single IMAP mailbox for an unrelated reason. But I am concerned about user storage issues at least..
Currently my POP users have a max mbox size but that is how much it will store before they "come get it". I ensure they have at least one device set to delete from the server periodically (i.e. 1 week). So users with multiple devices get their mail duplciated on their devices and at least one devices keeps the server emptied so their mbox does not hit any limits.
As I understand it, when using IMAP, the user's messages are perpetually stored on the server. So if the user is not moving their messages from their inbox to some other place (e.g. a different Outlook pst file), that all the messages will remain on the server. All messages are "synced" in the user's inbox or other user created folders. Is that right? This would pretty quickly cripple those normal non-computer types that have years of emails in their inboxes. And I don't have the storage space to allow storing years of their mail. I also understand if they exceed their storage limit, mail starts bouncing which is what's I'd expect.
At first I started googling how to expire the old mail, but then realized that won't work because if I delete on the server, I understand that will delete on their devices, too.
How do other's deal with this? Am I misunderstanding something? As popular as it seems to be, these seem like big hurdles to overcome. Appreciate some help for a IMAP noob.
Thanks, Scott