System errors

Ops inbox for Salesforce, ZingHR, credentials, schedulers, health, and leave integration failures — with clear cause and next steps.

Scroll to Flowcharts. Scenario: Triage a system error and resolve it.

Ops inbox

Why Salesforce, ZingHR, or health failed — and what to do

System errors is the role-gated ops inbox for integration and platform failures — Salesforce, ZingHR, credential reload, schedulers, host health, and leave-integration DLQ — with human-readable cause and a suggested next step.

Admin → System errors Route /admin/system-errors Org flag systemErrorsEnabled HR Admin · Org Admin

The job of this console

Scan open and critical failures, open a row for Why this failed and What to do, follow the deep link to Connections or Jobs, then Acknowledge or Mark resolved after you fix the root cause. This is an operator ledger, not a second source of truth.

Seat at the table

ADMIN and HR_ADMIN personas. Enabled by default; set adminModules.systemErrorsEnabled to false on a role to hide it. Org can turn the master off with systemErrorsAdminEnabled.

The working rhythm

Filter by status and source → open a row → read summary and suggested action → fix credentials or retry the job → Acknowledge or Resolve. Repeat when the same fingerprint bumps occurrence count.

Moments that bring you here

PO import stuck after Salesforce auth failure. Leave sync quiet while ZingHR is down. Connections health red. Secrets file write failed on reload. Scheduler catch for SALESFORCE or ZINGHR_LEAVES.

ⓘ

Ledger, not the owning system

Sync logs, health probes, and leave DLQ remain the technical sources. System Errors stores masked context only — never passwords or tokens. Fix the owning console (Connections, Jobs, Leave integrations), then resolve the event here.

Flowcharts

1 — Triage an open failure

Filter open errors, open a row, fix the owning system, then acknowledge or resolve.

flowchart TD
  A[Open System errors] --> B[Scan Open / Critical KPIs]
  B --> C[Filter source or status]
  C --> D[Open a row]
  D --> E[Read Why / What to do]
  E --> F[Fix Connections or Jobs]
  F --> G[Acknowledge or Resolve]

1
Open System errorsAdmin → System errors.
2
Scan KPIsOpen, Critical, and Last 24h set the priority.
3
FilterStatus and source chips narrow the register.
4
Read the drawerWhy this failed plus suggested action and deep link.
5
Fix and closeUpdate credentials or retry, then Acknowledge / Resolve.

Overview

This console is the operator-facing ledger over integration and platform failures. Events dedupe by fingerprint (occurrence count bumps on repeat). Sync logs, health probes, and leave DLQ remain the technical sources — System Errors stores masked summaries and suggested actions only.

How to use

1. Default on; role can disable

Unset role flags stay enabled. Set adminModules.systemErrorsEnabled to false to hide the menu for a role.

2. Never store secrets

detail_json is sanitized — passwords and tokens are masked.

3. Fix the owner first

Use the deep link to Connections, Jobs, or leave integrations, then resolve the event.

4. Acknowledge vs resolve

Acknowledge means you are working it; Resolve means the failure is handled.