How do mail search / backup programs work with iCloud?

I’m wondering about using something like InfoClick or MailSteward for backup of and search among my emails. All of my mail accounts use imap (iCloud, FastMail). As I understand it, which is probably not all that well , if a mail is present on one of the connected machines then it is in a hidden folder somewhere. Can either of these programs - or any others - find, catalogue, and backup all mail?

Many thanks

IMAP mail exists exists on the Server as well as on every client that subscribes to that account in a not very well hidden folder in the users’ Library/Mail folder, but not easily found in a series of nested folders after that. I do use InfoClick for searching as it allows me to quickly search by sender and/or by content, but it builds it’s own database to do that, which initially take some time. I do not consider it to be any kind of backup, however.

It’s not clear to me why you need a backup specifically for email. It’s true that if you fully delete an IMAP message it will eventually vanish from the server and all clients when they sync, but using TimeMachine or some other backup or cloning service will backup those deleted messages for whatever amount of time they are valid for.

Thank you Al.

I conclude from your message that InfoClick can find all mail in the library folders, that all mail is to be found in those folders, and thus InfoClick can and does index all mail.

I tried InfoClick a while back (maybe a year or two?) and abandoned it as it appeared (in my experience) only to index email in Mail “on this mac” but not that in the cloud. It was to address this that I asked the question.

As for the backup, I’m looking to address a couple of things. One is that I have all of my accounts on one machine and only a subset of accounts on the others, so my thinking was to create an SQL database of all mail that I can use on any machine should the need arise. Another is - as you hint - the longevity of TM and clone backups; I’m pretty sure that old mail has already fallen off the end of both clones and TM backups. Not only that, but if I ever need to access those old mails on eg TM backups it’s a lot more effort than simply looking them up on an SQL database.

It depends on your mail client app. Most apps will cache recently-read messages so you don’t need to re-download it every time you look at it (and so you can read mail when off-line).

Some apps may purge their cache of old messages, re-downloading them only when you choose to view them in the future, in order to save space on your hard drive.

But I wouldn’t consider that a backup, as such, because the cached data will be deleted when the message is deleted from the server (and the app syncs its folder).

If you need to keep a local copy, you should copy or move the message(s) in question to a local folder. The contents of local folders aren’t synced with any server and will therefore not be affected by what happens on-line. The mechanism for creating a local folder and copying/moving messages (whether manually or automated) will depend on what app you use to read the mail.

FWIW, on my Mac, I usually read mail using the provider’s web interface. I use Apple’s Mail app on my iPhone, which accesses them via IMAP. When I want to use an app on my Mac, I use Mozilla Thunderbird.

I don’t generally retain messages for a long time. Most get deleted (moved to the server’s Trash folder until the server purges them) after I read them, or after the circumstance of the message is no longer relevant (usually a few days or a few weeks later). For those I want to keep for a long time, I create folders on the IMAP server and move the messages there. For those few that I want to keep locally, I have created local folders in Thunderbird, and I move those messages there.

My Thunderbird installation’s local mail folders are backed up along with all my other documents. Everything else remains on the servers. This hasn’t ever been a problem, although it is theoretically possible that a system glitch could cause me to lose it all.

1 Like

Thank you for your thoughts, David.

I’ve now discovered (one of?) the reasons I saw limited scanning of email in my previous evaluation of InfoClick. The privacy settings got in the way - which I should have thought to check earlier.

The replies from Al and David helped me “think different”. Thank you again.

I prefer to not use email as a filing system. Any emails I want to save I export as a PDF and include into my paperless filing system. This keeps related documents together instead of having some in emails and others as documents.

Anybody with business related emails should make sure that you are in compliance with your document retention policies. Failure to do so can put you in legal jeopardy. I advise consulting your legal counsel about it.

Your point is well taken, Glen. Although this is not commercially important stuff, I too would like a safer repository. Which is one of the reasons I was looking at MailSteward.

Why not just drag the email to a finder folder without going through PDF conversion. It’s then available as long as the mail program is.
I’ve started doing that and am interested in why that’s not a good idea instead of keeping mail in mail folders.


Therein lies the problem. PDF, especially PDF/A is a much more stable document format than email. All my scanned documents are in PDF as well as a lot of saved documents received as attachments.

This depends entirely on the format your mail app is using.

The old Berkeley mbox file format is still commonly used and is a very stable format. It is pretty much the raw text data (with its binary attachments encoded as text-encoded blocks) that SMTP servers exchange as they deliver mail. An mbox file is just a sequence of these messages concatenated together with one special text line (a From header, without a colon) to separate the messages.

This format is nice because it should be easily importable into any e-mail software. And it preserves the entire message, including alternate renderings that might be present (e.g. some messages include both plain-text and HTML versions of the content) and all of the header lines generated/relayed during the delivery process.

On the other hand, PDF is going to contain the result of rendering the mail message for viewing/printing. It will discard any data that doesn’t get displayed/printed by the application you use to generate it.

Of course, mail apps may also save messages in proprietary formats (e.g. Outlook’s “.msg” files) that other applications may have a hard time reading. I agree that you probably don’t want to be saving those if you think you may need to share them with people using other apps.