Category: Monitoring & Alerts · Topic: Communication Insights · Applies to: All Users
Overview
The Insights page is your centralized hub for notifications, alerts, and system insights. When a communication insight appears — for an entire system, a specific inverter, or a string — it does not always mean there is a real outage. This article walks you through how to investigate the most common communication insight types, identify the root cause, and resolve or escalate each issue correctly.
Note: Insights are data-driven and fully automated. You cannot create or delete them manually — they open when the data signals an issue and close automatically once the data returns to normal.
Troubleshooting a System-Level Communication Insight
A communication insight for an entire system can appear even when the vendor platform (e.g., SolarEdge, Meteo, CPS, Also Energy, eGauge) shows data. Work through the steps below in order until you identify the cause.
Possible causes:
The system is communicating, but data arrives too early or too late relative to Ensights' alert window.
Live data (15–60 min intervals) is delayed or missing, while daily-level data pulls correctly.
API credentials or keys for this vendor were changed and not updated in Ensights.
The vendor's API has blocked Ensights due to API call rate limits.
The vendor has updated their platform or API.
Step 1: Check whether other systems from the same vendor source are also affected. On the Insights page, use the Group by filter to see if multiple systems share the same communication insight. If many or all systems from the same vendor are offline simultaneously, the cause is almost certainly an API issue — changed credentials, a rate-limit block, or a vendor platform update.
a. If credentials or API tokens were recently changed, contact Ensights support and provide the new credentials.
b. If credentials have not changed, contact Ensights support to investigate. For Enphase and APsystems, check API call statistics in the vendor platform before escalating — include those statistics in your report.
Note: If a vendor API change is the root cause, the Ensights team will update the platform connection internally.
Step 2: If only one system is affected, check the system chart on a daily level. Open the system chart and confirm whether any data is present for today. Note the timestamp of the most recent data point.
Step 3: Evaluate the timestamp.
If the timestamp is earlier than the current time (e.g., you check at 10:00 AM and the latest data shows 11:00 AM), the system's local clock is likely running ahead — common after daylight saving time shifts. Adjust the clock in the vendor platform or contact vendor/O&M support.
If the timestamp is 60 or more minutes behind current time, you have two options:
a. Contact your O&M or vendor support to correct the delay.
b. Extend the alert validation window in Ensights: navigate to the Pvsystem page → click the three-dot menu on the relevant system → click Edit → go to the Advanced tab → set a longer Alert Validation Window. Ensights will then check the full window before triggering an alert.
Step 4: If there is no data in the daily chart at all, check whether daily production appears under the Monthly resolution. Switch the chart to Monthly view. If you can see daily production for past days, the issue is specific to live (15–60 min) data.
Note: If live data exists in the vendor platform but not in Ensights, contact Ensights support for further investigation.
Step 5: If none of the above steps resolve the insight, escalate to Ensights. The Ensights team will investigate the API connection internally. Resolution may take 2–4 weeks if a vendor API change is involved.
Tip: Use the 'Group by' filter on the Insights page to quickly identify whether the communication issue affects one system or many at the same time — this is the fastest way to distinguish a single-system timing issue from a vendor-wide API problem.
Troubleshooting an Inverter-Level Communication Insight
Follow these steps when a communication insight points to a specific inverter while the vendor platform shows data for that inverter.
Possible causes:
The inverter is communicating, but data is arriving too early or too late.
The inverter was physically replaced and the new unit has a different serial number.
Step 1: Check the system chart on a daily level for the affected inverter. Confirm whether data is present and note the most recent timestamp.
Step 2: Evaluate the timestamp using the same logic as a system-level insight.
If the timestamp is ahead of the current time — likely a clock sync issue from a daylight saving time shift. Contact vendor or O&M support to correct the inverter clock. If the timestamp is 60 or more minutes behind, you have two options:
a. Contact O&M or vendor support to fix the delay.
b. Set a communication delay in Ensights: navigate to the Pvsystem page → click the relevant system → scroll to the Inverter table → click the three-dot menu on the affected inverter → click Edit → select the appropriate Communication Delay. Ensights will check the full window before alerting.
Step 3: If the inverter was recently replaced, sync the system to pull the new serial number. Navigate to the Pvsystem page → click the three-dot menu on the relevant system → click Sync and wait 5–15 minutes. Refresh the inverter table — a new inverter entry should appear.
Note: After syncing, you must handle the old inverter entry. See the options below.
Step 4: Handle the old inverter entry.
a. To remove it permanently: click the three-dot menu on the old inverter → click Delete.
b. To keep its historical data: click Edit → enable Suspend Alerts. An orange circle icon will appear next to the inverter to indicate alerts are suspended. Rename the inverter to include '(old)' so teammates can identify it easily.
Step 5: If sync did not detect a new inverter or the steps above did not resolve the insight, contact Ensights support.
Important: Deleting an inverter permanently removes its historical data from Ensights. If you need to preserve past performance records, use 'Suspend Alerts' instead of Delete.
Troubleshooting a String-Level Communication Insight
A 'No current in string' insight can appear for strings that are intentionally empty — not connected in the field — but still visible on the vendor platform showing 0 current.
Step 1: Confirm the string is genuinely unconnected. Verify with your field records or O&M team that the string is not supposed to carry current and is not a wiring fault.
Step 2: Mark the string as not connected in Ensights. Navigate to the system on the Pvsystems page → open the Insights tab → click the three-dot menu on the relevant insight → click Mark as Not Connected. The string will be removed from monitoring and will no longer generate alerts. This may take a few minutes to take effect.
Important: This action is irreversible. Only mark a string as not connected if you are 100% certain it is not supposed to carry current.
Tip: To dismiss multiple string insights at once: set Group by to None on the Insights page → click the three-dot menu above the table → select Multiple Selection → choose all the relevant string alerts → click Mark as Not Connected. This same action can also be performed directly from the Insights page.
Quick Reference: Communication Insight Types and Next Steps
Use this table to quickly identify the insight type and where to start your investigation.
Insight Type | Scope | First Step |
Communication error | Entire system | Check if other systems from the same vendor are also affected — if yes, likely an API issue |
Communication error | Inverter | Check daily chart timestamp for that inverter |
No current in string | String | Confirm string is intentionally unconnected in the field |
API or vendor platform mismatch | Multiple systems | Contact Ensights support with new credentials or API tokens |