Aurora Hub — Operations Notes

If the site feels stuck or you've just changed something, this is the one command that fixes most things.

sudo systemctl restart task-app

Then confirm it actually came back:

sudo systemctl status task-app
curl -I https://hub.auroracomputercare.com/login

You're looking for Active: active (running) and HTTP/1.1 200 OK.

Everything the app prints — every page load, every scheduled email attempt — goes to the system journal.

# last 50 lines
sudo journalctl -u task-app -n 50

# live, updates as things happen
sudo journalctl -u task-app -f

# everything since midnight
sudo journalctl -u task-app --since today

Email failures specifically also get their own file, which doesn't roll over the way the journal eventually does: data/mail-failures.log.

This is the one command that answers it — every scheduled job (the four check-ins and the nightly backup) writes one line the moment it finishes, whether it worked or not:

sudo journalctl -u task-app --since today | grep scheduler

A healthy day reads like this:

Those timestamps are UTC, so they'll read a few hours later than the 6am/7am/noon/6pm Denver time you actually see the email land. Silence instead of a line means the job never ran at all, which is worth noticing on its own.

Schedule: every night at 3:00am Denver time, automatically — no separate job to maintain, it's the same scheduler as the emails.
Kept for: 30 days, then quietly deleted. At this app's size that's still under a megabyte total, so there was no reason to keep less.

ls -la ~/projects/task-app/backups/

Want one right now, before trying something risky? Run it by hand:

cd ~/projects/task-app
node scripts/backup.js

This replaces the live list with an older one. Anything added or checked off after that backup's date is lost — check the date before you commit.

1. Stop the app so nothing writes to the database mid-restore:
   step 1Copy

   sudo systemctl stop task-app

2. Save the current database aside, just in case:
   step 2Copy

   cp ~/projects/task-app/data/tasks.db \
      ~/projects/task-app/data/tasks.db.before-restore-$(date +%Y%m%d-%H%M%S)

3. Copy the backup you want over the live database — swap in the date you actually need:
   step 3Copy

   cp ~/projects/task-app/backups/tasks-2026-09-04.db \
      ~/projects/task-app/data/tasks.db

4. Start it back up and confirm it's serving the restored data:
   step 4Copy

   sudo systemctl start task-app
   curl -I https://hub.auroracomputercare.com/login

It lives in one file. Edit it, then restart:

sudo nano ~/projects/task-app/.env
# edit the PASSCODE= line, then Ctrl+O, Enter, Ctrl+X to save
sudo systemctl restart task-app

Anyone already signed in stays signed in — only new logins need the new code.

Run the full test suite, then the automated browser check that loads the real app and verifies nothing looks broken on the very first screen:

cd ~/projects/task-app
npm test
npm run smoke-test

The cache the app uses for offline/installed use updates itself automatically whenever a file changes — there's no manual version number to remember to bump.