Is he using cached mode?

Could this user have a cached DNS record for their domain on the machine in question by chance? Since its happening during outages I wonder if the user's Outlook client is failing over to that cached DNS record and landing on an old server. 

Outlook has had a maximum file size for PST and OST file which defaults to something like 50GB. This setting, and a few related settings, can only be modified in the registry. If a profile's OST or PST file reaches or exceeds that limit, it can cause issues in Outlook's behavior. Since you indicated that the account size in SmarterMail is around 40GB, it's possible that one of the Outlook files is at that limit.

Here's a link to a page with more information on the settings in question: https://learn.microsoft.com/en-us/microsoft-365-apps/outlook/data-files/configure-size-limit-outlook-data-files

Good points! Yes, he's using cached mode. The outage was not at their office but at our hosting site. The DNS record shouldn't change though even if the server is offline, whether due to our Internet outage/SM server issue.

But, to answer your point, Kyle, this specific customer has an AD design from around 2001 (since upgraded multiple times to current DCs and AD functional levels) but their AD domain is formatted as "acme.com" (matching their public name) rather than "acme.local" so they have split DNS as a result. Regardless, only one PC in the AD domain is problematic, which makes me wonder what the heck is going on. Of course, the offending PC belongs to the company manager (just my luck) and no one else is exhibiting the problem... I really don't expect deep analysis of the potential causes from the community, but if anyone has had a similar incident along with solution, I'd appreciate any info to ponder.

Creating a new profile and beginning a new mailbox sync "solves" the issue but it doesn't get at the underlying cause, even if it is only one PC. This specific situation only contributes to my slowly-evolving Luddite conversion over the decades, LOL. Thanks, guys.

That makes sense Matthew and the manager being the affected user definitely makes things trickier! One thing you could try is installing Fiddler Web Debugger classic on their machine and configuring it to decrypt HTTPS (in the options area.) The next time this happens the manager can open up Fiddler and look for red entries to see who/what its trying to communicate with when it fails. He can export those results as an SAZ file as well and you can send that our way to take a look as well. 

Andrew's suggestion is another good one too as I've personally seen large mailboxes blow up Outlook. One idea I've had some success in convincing users of is to move mail oder than X number of years to an archive account of sorts. If their domain is domain.com, you can create a second domain called domain-archive.com and keep their oldest email in there. Should they ever need it its just a webmail login (or sync) away, but it keeps it out of their main account and therefore out of the Outlook profile. If you have older slower drives you can put the archive domain there too and save your SSD space for the mail they need regularly.