Breadcrumbs

HCMS: Core System Diagnostics

1. Overview

This KB defines the minimum diagnostics required when HCMS appears slow, unresponsive, or returns outdated data. It applies to customer‑managed, partner‑managed, and Censhare‑hosted environments. It clarifies which checks can be performed by application teams and which require Infra/DevOps involvement.

2. Symptoms

  • Slow or unresponsive HCMS API

  • Outdated or delayed API responses

  • Intermittent 502/503 errors

  • Delayed propagation of data changes

  • Suspected load‑related instability

3. Common Causes

  • Backend resource saturation on the Censhare Server

  • Unstable or restarting HCMS satellite processes

  • High request volume from integrations or tools

  • Replication delays between Censhare Core and HCMS

  • Incorrect or partially applied schema/configuration deployments

  • Recent high‑impact operations such as bulk imports or migrations

4. Diagnostics

4.1 Customer/Partner

  • Review whether integrations generate heavy request traffic (frequent polling, parallel calls, unthrottled batch operations).

  • Verify whether updated data appears in Censhare Core and compare it with HCMS API responses.

  • Check deployed schemas via the HCMS schema endpoint and confirm the expected schema is active.

  • Identify any high‑impact workloads active during the issue period (migrations, bulk writes, sync tasks).

4.2 Infra/DevOps

  • Check backend load on the Censhare Server (CPU trends, memory usage, disk I/O, long‑running operations).

  • Verify satellite stability (restarts, CrashLoopBackOff, OOMKilled, throttling).

  • Review ingress/load balancer health and confirm satellites are available.

  • Check replication status if delays between Censhare Core and HCMS are suspected.

  • Validate that required configuration components (e.g., image service) are deployed and healthy.

4.3 Censhare Support

  • Perform backend and satellite health validation on hosted environments.

  • Confirm replication status and investigate replication lag.

  • Review system-side stability indicators to identify persistent or structural issues.

5. Mitigation

  • Reduce or throttle high‑frequency or parallel API calls.

  • Pause or reschedule heavy workloads during peak usage times.

  • Correct or reapply schema/configuration deployments if inconsistencies are found.

  • Request replication checks when data appears outdated despite being updated in Censhare Core.

  • Address resource bottlenecks identified on backend or satellite nodes.

6. Best Practices

  • Throttle automated integrations and batch large write operations.

  • Avoid heavy tasks during business‑critical time windows.

  • Validate schema deployments before activating dependent functionality.

  • Maintain sufficient resource headroom in backend and satellite infrastructure.

  • Ensure monitoring and alerting are in place for load indicators and satellite stability.

7. When to Contact Support

Contact Censhare Support when:

  • Backend load appears normal but HCMS performance remains affected.

  • Satellites are stable yet HCMS still returns outdated data.

  • No large workloads were active during the issue window.

  • Schema and configuration appear correct.

  • The issue persists across multiple requests or for an extended timeframe.

8. Information Required

Provide the following when opening a support request:

  • Timeframe when the issue occurred

  • Type of workload active (read/write/migration/sync)

  • Description of observed symptoms

  • Whether the updated data appears correctly in Censhare Core

  • Recent schema or configuration changes

  • Any high‑impact operations occurring at the time

  • Actions already taken based on this KB