This is the right first move. A healthy container in docker ps only means the process is running, not that the app has finished starting, unless you wrote a real health check.
The real question is not the hours, it is whether your redundancy and backup plan assumes these drives will fail. If it does, 41000 hours is a bargain. If your plan is that they must not fail, no drive is cheap enough.
Eleven weeks in summer is honestly not unusual in my field and often has nothing to do with you. Field season, grant deadlines, a sick parent. Doesn't make it acceptable, but it should lower your anxiety a bit while you apply the practical fixes people are describing.
Also worth checking whether the destination dataset has compression or encryption doing something unexpected, and whether the source is a spinning disk that is itself slowing. Half the time the bottleneck is on the machine nobody suspects.
Also worth noticing that rereading three papers and moving a figure is not nothing. It looks like zero output because it leaves no trace in the word count.
The database dumps, near certainly. A full dump written as a fresh file every night deduplicates badly because compression changes the byte layout of the whole file even when 99 percent of the data is identical. If they are compressed dumps, that is your 400GB. Two fixes: dump uncompressed and let restic handle the compression and dedup, or use a format that produces stable output. People are always surprised how much a 10GB nightly dump costs over a month.
There it is. Switch to plain dumps and the same month will cost you a few gigabytes. Also check whether gzip is set to include a timestamp in the header, because that alone can defeat dedup on otherwise identical files.
Technically true and tactically terrible with a person who signs your defense form. There is a version of no that sounds like scoping, and that is the one to use.
Did not know the business manager was a distinct person from payroll. Emailing her tomorrow.