Browse
On this page

Owner controls, background jobs, and the finder tools under Settings > System

Owner Controls, Data Retention, Integration Sync, Scheduled Jobs, Background Tasks, Admin Tools and All Settings all live on Settings > System, plus two doorway pages that point to Integrations and Logs.

You need An account with view access to Settings. Owner Name and Owner Email on Owner Controls save only for the tenant's own owner account or an internal Super Admin; everyone else can read them but not change them.

What you will have

  • Find the owner-only Owner Name, Owner Email and System Email fields, and see why they are locked to the owner.
  • Read Data Retention to see exactly how many days each kind of log, event and chat is kept before it purges.
  • Read Integration Sync's device activity window, warranty lookup budgets and renewal warnings, and see that PSA Write-Back is permanently off.
  • See how often the platform's own background jobs actually run, on Scheduled Jobs.
  • Watch a job you started, like writing a year of security awareness training, on Background Tasks, and see everyone else's too.
  • Find where to add or edit the help text that shows in a page's own question-mark drawer, on Admin Tools.
  • Read System Version to see the running build, and confirm Automatic Updates is off.
  • Search every setting on the platform by name from All Settings, and jump straight to it.
  • Use the Integrations and System Logs doorway pages to find those two areas from inside Settings.

Why it works this way

PSA Write-Back is not a switch you can flip on this tenant. The 2026-08-29 'WE ARE THE PSA' ruling permanently retired it, so Integration Sync always shows it Off and view-only; a connected PSA is now a read-only migration mirror you disconnect from Admin > Integrations > PSA when you are ready to cut over.

System Version's Release Channel normally reads a channel name like Canary, Dev or Stable. This demo tenant is deliberately pinned to one build for consistent screenshots, so its Release Channel field shows that build's short commit id instead of a channel name - your own portal reads a channel name here.

A Help Entry's navPath has to match a page's help path exactly, not as a prefix. An entry saved at admin.settings.sla shows only on the page whose own help path is exactly admin.settings.sla; it never rolls down to a sub-page like admin.settings.sla.critical. Covering several pages with one topic means adding a separate entry for each navPath.

Scheduled Jobs and Background Tasks both answer 'what is running', but they are opposite halves of it on purpose. Scheduled Jobs is the system's own clockwork - cadences tuned in code, running whether anyone is here or not. Background Tasks is the ledger of jobs a person clicked to start; nothing on it runs on its own, and there is nothing on it to configure.

Integration Sync's own numbers are windows and lookaheads - how far back or how far ahead each sync looks - not how often it runs. The tick interval itself (clients every 6 hours, tickets every 1 minute) is set on Scheduled Jobs; that page's own cadence badges show the human-readable schedule, and hovering one reveals the underlying crontab expression.

Steps

  1. Open Settings > System > Owner Controls.

    Owner Name and Owner Email are both tagged (owner only): only the tenant's own owner account, or an internal Super Admin, can save changes to them, even though any admin with Settings access can view the page. Owner Email is the account that auto-promotes to Super Admin on first sign-in, so treat it like a domain-transfer credential. System Email is the sender and owner address for records the system creates on its own, such as users and tickets that arrive from integration syncs.

    Open Settings > System > Owner Controls.
  2. Open Data Retention to see how long each kind of record is kept.

    Every field here is set automatically and view only. Info logs keep 1 day, warn logs 2 days, error logs 30 days, the processed event ledger 30 days, rule-engine decision rows 90 days, the outbound email log 180 days, read notifications 90 days, and unread notifications 180 days - longer than the read window, so an alert nobody has seen yet survives. Archived Elise chats default to 0, which means keep forever; active chats are never purged either way. Error logs, the event ledger, unprocessed events and emails still in flight are held longer, or never purged, because the AI diagnostic engine and support both rely on them to explain what happened later.

    Open Data Retention to see how long each kind of record is kept.
  3. Open Integration Sync and scroll to the Cadence card at the bottom.

    The fields on this page - Device Activity Window, Device Aging Threshold, the per-vendor warranty lookup budgets, Days Before License Renewal and MS Renewal Warning Days - are windows and lookaheads, how far back or ahead each sync looks. PSA Write-Back on this same page shows Off and View only: it is permanently retired, always off, and can no longer be turned on. The card at the bottom says the tick interval itself lives on Scheduled Jobs (clients sync every 6 hours, tickets every 1 minute); adjust cadence there, and the windows here to tune false-positive churn.

    Open Integration Sync and scroll to the Cadence card at the bottom.
  4. Open Scheduled Jobs to see every cron job's own cadence.

    Each row names the job, what it does, its plain-English cadence badge such as Every 5 minutes or Daily at 4:00 AM, a Run Now button rate-limited to once a minute, and an On/Off toggle that pauses the job entirely. Hovering a cadence badge shows the exact crontab expression behind it. The page's own intro says cadences are tuned in code for vendor API quotas and cost, and that the editor list and the real job runner can drift - some jobs run on hardcoded defaults with no editor row, and some editor rows point at jobs the runner does not register - so treat the list as a guide, not a guarantee.

    Open Scheduled Jobs to see every cron job's own cadence.
  5. Open Background Tasks to see what you and your team have started.

    This ledger is the opposite of Scheduled Jobs: nothing on it runs on its own and there is nothing here to configure, only jobs a person clicked to start that keep running after the page returns - the real example on this tenant is 'Write annual security awareness training,' shown here Finished for 2026 Security Awareness Training. Each row shows the task, what it was for, who started it, its status, when it started and finished, and a Result column. The card opens on your own tasks, because that is usually what brought you here.

    Note: Read the Result column first if a task fails - it is empty here because nothing has failed, but on a real failure it carries the actual reason plus an incident number when one exists, not a generic apology. Some failures are worth retrying and some will fail identically every time; nothing is ever half-saved, since a failed job leaves what it was working on exactly as it found it.
    Open Background Tasks to see what you and your team have started.
  6. Click Show everyone's to widen the list past your own tasks.

    The card's title changes to Everyone's background tasks and the button flips to Show only mine. This is the view for working out why a colleague's training or report never appeared. Seeing everyone's tasks is its own permission, separate from ordinary Settings access, so being able to change a setting does not by itself let someone read what the whole team has been running.

    Click Show everyone's to widen the list past your own tasks.
  7. Open Admin Tools and scroll to Help Entries.

    Above Help Entries, the same page also holds the Admin Navigation Tree (drag rows to reorder the sidebar; order saves instantly, and field edits save from the editor panel on the right). Help Entries itself lists every HelpEntry with its Question, the exact Page navPath it is pinned to, whether it is Active, and its sort order, with a search box and a New Entry button. Each entry's navPath has to match a page's help path exactly - it never rolls down from a page to that page's own sub-pages - so covering several pages with one topic takes a separate entry per navPath.

    Open Admin Tools and scroll to Help Entries.
  8. Scroll down to System Version.

    This card shows the running Docker image's short commit id as Current Version, its Build Date, and a Release Channel field - normally a channel name like Canary, Dev or Stable, though this demo tenant is deliberately pinned to one build so its Release Channel shows that build's id instead. An Up to Date row confirms the instance is Current, and Automatic Updates sits below it, off. Updates do not happen from this page: deployment is handled by the platform's own image pipeline, and this toggle is inert legacy UI.

    Scroll down to System Version.
  9. Open All Settings and expand a category.

    The search box searches every registered setting across the platform, and the line beside it counts what the search is showing against the whole set, 191 of 191 here with nothing typed. Below it, a chip per category (Billing, Clients, Email, General, Integrations, rmm, Security, SLA, Tickets) filters the list to that category; the chips carry no counts of their own. Expanding a category, Email here, lists each setting's name, a plain description of what it does, its current value, and an Edit link that jumps straight to the page that edits it. There is no automated check that a relocated setting's Edit link still points at the right field, so if a setting moves without this entry being updated, the link can land on a page that no longer edits it.

    Open All Settings and expand a category.
  10. Open Integrations to see why it lives outside Settings.

    This doorway page explains that connections to outside systems - PSA, RMM, identity, accounting - act like their own marketplace of vendor tiles with live connection state, so they live in the Integrations area instead of a Settings page. Its four steps: open Integrations, pick the provider you want (QuickBooks, Microsoft, your RMM), enter or authorize its credentials, then enable the connection and confirm it shows as connected. The Open Integrations button jumps straight there.

    Open Integrations to see why it lives outside Settings.
  11. Open System Logs to see why it also lives outside Settings.

    This doorway explains that logs are an operational record you read and filter, not a setting you configure, so each stream is kept separate in its own Logs area. The card's own five steps are: open Logs, choose a stream, read that the assistant's own action log is not one of them, filter by date, user or source, then open an entry to see its full detail. The stream list named on this card is Audit, Email, System, Operations or Sessions. The assistant's action log lives under its own Action History tab instead.

    Open System Logs to see why it also lives outside Settings.
  12. You do not have to stay on Background Tasks while something runs.

    Once a task is listed it is running on the server, not in your browser, so closing the tab, going to lunch, or signing in from another machine makes no difference. The page polls for updates every 5 seconds only while at least one of your tasks is still running, and stops polling once everything is finished, so a completed ledger costs nothing to leave open. The only thing that can interrupt a task is the portal itself restarting mid-run; a task caught by that is marked failed with a note saying so, rather than sitting on Running forever.

If it did not work

  • If Owner Name or Owner Email will not save, that is by design - only the owner account itself or a Super Admin can change them.
  • If Integration Sync's PSA Write-Back looks stuck on Off with no way to enable it, that is not a bug - write-back was permanently retired and the field is now view only.
  • If a task you know a colleague started never appears on Background Tasks, you are looking at My background tasks - click Show everyone's.

Questions this page answers

Why is Owner Email marked sensitive?

Owner Email is the single account that auto-promotes to Super Admin on first login, so it is the seed of trust for the whole portal. Sensitive and owner-tier settings, including Owner Email itself, can only be changed by the owner account or by a Super Admin, so keep that circle small. Treat it like a domain transfer credential, losing it means re-establishing trust at the Super Admin layer.

What is the Data Retention page for?

This page sets how long each kind of log, event, and chat is kept. Nightly cleanup deletes them after that. It covers info logs, warn logs, error logs, the event ledger, the email log, notifications, and archived AI chats.

Why are some of these kept much longer than others?

Error logs, the event ledger, unprocessed events, and emails still in flight are held longer, or never purged. That is on purpose. The AI diagnostic engine and support both rely on them to explain what happened later.

How often does sync actually run?

The values here are windows + lookaheads, how far back to look on each tick. The tick interval itself (clients every 6 h, tickets every 1 min) lives in Settings → System → Scheduled Jobs. Adjust the cadence there if your API budgets need more headroom; adjust the windows here to tune false-positive churn.

What lives on Cron Schedules?

The schedule and on/off state of each cron job the portal runs (PSA sync, ticket-review auto-close, recurring tickets, billing rollups, and so on), with per-job toggles and a crontab-syntax schedule. Note that the editor list and the actual job runner can drift: some jobs run on hardcoded defaults because they have no editor row, and some editor rows point at jobs the runner does not register, so treat this list as a guide, not a guarantee.

What is this page for?

It lists the jobs you and your team started that keep running after the page comes back. Some work takes minutes rather than seconds - writing a whole annual security awareness training with AI is the current example - and making you sit on a loading spinner for three minutes is both unpleasant and fragile, because the connection times out long before the work does. So the click returns immediately, the work carries on, and this is where you come to find out how it went.

How is this different from Scheduled Jobs?

Scheduled Jobs is the system’s own clockwork - the syncs, sweeps and reminders that run on a cadence whether anyone is here or not, and where you set that cadence. This page is the opposite: nothing on it runs on its own, and there is nothing here to configure. Every row is something a person clicked. Keeping them apart is deliberate - if the system’s hourly work were mixed in, the one question this list answers, "did the thing I started finish?", would be buried under noise.

One of my tasks failed - what do I do?

Read the Result column first. It carries the actual reason, not a generic apology, because the reason decides what to do: some failures are worth retrying immediately and some will fail identically every time until something is changed (an AI answer running past its length limit is the classic one - retrying that is the one thing guaranteed not to work). Where a failure has an incident number, the full technical detail is under Admin › Logs against that number. Nothing was half-saved: a job that fails leaves the thing it was working on exactly as it found it.

Can I see what other people started?

Yes - the list opens on your own tasks, because that is usually what you came for, and the button in the card header switches it to everyone’s. That is the view for working out why a colleague’s training never appeared. Access to the page as a whole is its own permission, separate from general Settings access, so being able to change a setting does not by itself let someone read what the whole team has been running.

Do I have to stay on the page while something runs?

No. Once a task is listed it is running on the server, not in your browser - close the tab, go to lunch, sign in from another machine, it makes no difference. The page you started from updates itself if you happen to still be on it, and this list is here for when you are not. The only thing that can interrupt a task is the portal itself restarting, and a task caught by that is marked failed with a note saying so rather than sitting on "Running" forever.

Can one entry show on multiple pages?

Not directly, each HelpEntry has one navPath and the drawer matches it EXACTLY, not by prefix. An entry saved at admin.settings.sla shows only on the page whose help path is exactly admin.settings.sla; it does not roll down to sub-pages like admin.settings.sla.critical. To cover several pages with one topic, add a separate entry for each navPath.

What does System Version show?

The running Docker image SHA, the channel (canary / dev / stable), and whether a newer image exists on the channel. Updates do NOT happen from this page: deployment is handled by the platform's image pipeline, not by these controls. The toggles here are inert legacy UI.

What is the All Settings tab?

Search across every registered setting and every page where it can be edited (the Edit links). Useful for hunting down where a particular setting lives when its home isn't obvious. One caveat: there is no automated check that an Edit link still points at the field, so if a setting is relocated without updating its entry here, the link can lead to a page that no longer edits it.

What is the Integrations link for?

This doorway explains why Integrations lives outside Settings. Your PSA, RMM, identity, and accounting links each show live status. They act like a small marketplace of vendor tiles. So they live in the Integrations area instead.

How do I connect a new integration?

Click "Open Integrations" on this page, or go straight to Integrations. Pick the provider you want, like QuickBooks, Microsoft, or your RMM. Enter or authorize its credentials. Then turn on the connection and check that it shows as connected.

What is the System Logs link for?

This doorway explains that logs are a record you read and filter, not a setting you configure. They live in their own Logs area, with each stream kept separate.

How do I find a specific log entry?

Click "Open Logs" on this page, or go straight to Logs. Choose a stream: Audit, Email, System, Operations, or Sessions. The assistant's own action log is under its Action History tab, not here. Filter by date, user, or source. Then open an entry to see its full detail.

Was this helpful?

Last validated 2026-09-18