Document ID: KB-EV4-API-01 | Product/Service: EloView 4 REST API | Audience: Integrators, Developers, Enterprise Customers
EloView 4 API Rate Limits and High-Volume Integration Best Practices
The EloView REST API enables enterprise customers and integration partners to programmatically retrieve device information, monitor fleet health, manage content, and deploy applications. This article outlines the EloView 4 API rate limits, authentication behaviors, and architectural best practices for reliable, high-volume API integrations.
💡 Quick Overview
EloView 4 does not apply individual customer-specific API limits. For optimal reliability, integrations should cache OAuth tokens, favor bulk/multi-device queries, limit concurrent threads, and implement exponential backoff with jitter for error handling.
1. EloView 4 API Rate Limits
EloView 4 does not enforce rate limits per individual OAuth credential, customer account, source IP address, or API endpoint. The production API base URL is:
https://secure-api.eloview.com
| Parameter | Limit / Capacity |
|---|---|
| Per credential, account, source IP, or endpoint | No specific rate limits applied. |
| AWS Platform Request Rate (Shared) | 10,000 requests/second (Shared globally by all customers). |
| AWS Platform Burst Capacity (Shared) | 5,000 requests (Shared globally by all customers). |
⚠️ Critical Integration Warning
The shared platform limits shown above are not reserved per-customer quotas. To avoid performance issues and protect platform stability, integrations must be designed for efficient, responsible resource usage.
2. OAuth Authentication and Token Management
Improper token handling is a common cause of avoidable API overhead. Please structure your authentication logic based on the following rules:
- Token Lifespan: Access tokens are valid for 24 hours (
expires_in = 86400). - Cache & Reuse: Cache and reuse access tokens for their entire valid lifespan. Do not generate a new token for every API request.
- Proactive Renewal: Renew the cached token shortly before expiry, or immediately upon receiving an
HTTP 401 Unauthorizedresponse. - Credentials vs. Capacity: While account administrators can create multiple OAuth credentials, utilizing multiple credentials does not increase your overall API capacity.
3. Monitoring Fleet and Device Status
When monitoring device health and active deployments at scale, optimize your query design to reduce network overhead:
- Routine Status Audits: Use bulk query endpoints—such as
GET /devicesorPUT /devices/search—to retrieve status records for multiple devices in a single call. Avoid pollingGET /devices/{serial}individually for each kiosk. - Polling Intervals (Idle): For routine health and status checks, polling intervals of every few minutes are sufficient.
- Active Deployment Monitoring: To track active installations, query
GET /device/{serial}/content/statusat an interval of 1 to 2 minutes. Stop polling once the deployment reaches a finalized state (Success or Failure).
4. Application Deployments to Multiple Devices
The POST /deployments endpoint supports targeting many devices simultaneously.
- Targeting Formats: Specify recipients in a single request payload using a list of serial numbers, device groups, or query criteria.
- The Empty Query Target: Leaving the target query blank will automatically target all devices in the account. Use this parameter with extreme care.
- Batch Restrictions: Although the endpoint does not enforce a rigid serial-number limit, payloads remain subject to your account's batch-size setting and the OAuth credential's batch-operation permissions.
- Over-Limit Responses: Exceeding your account limits may trigger a
400 BATCH_LIMIT_EXCEEDEDerror, or require manual verification ifPENDING_FOR_APPROVALis configured.
5. Concurrency, API Errors, and Retries
Integrations should maintain a conservative concurrency profile. Keep active parallel requests constrained to 5–10 concurrent connections. Use queues and batch processing to stream data rather than executing thousands of parallel endpoints.
Common API Responses & Actions
| API HTTP Status | Meaning & Recommended Action |
|---|---|
| HTTP 401 Unauthorized | Authentication failure (e.g., token expired or revoked). Verify authentication credentials and request a new token. |
| HTTP 429 Too Many Requests | Platform rate throttling. Implement retry logic using exponential backoff with randomized jitter. |
| 400 BATCH_LIMIT_EXCEEDED | The payload size exceeds permitted batch execution limits. Review account settings and partition the transaction. |
⚙️ Retry Logic Backoff Strategy
When handling HTTP 429 responses, configure your retry schedule progressively (e.g., 1s, 2s, 4s, 8s, 16s, 32s, up to 60s max) and apply a randomized delay (jitter) to prevent localized service floods. Note: Platform return of the 'Retry-After' header is unconfirmed.
6. Troubleshooting Unexpected API Errors
If your integration experiences persistent disruptions, run through this quick troubleshooting checklist:
- Validate the Base URL: Confirm you are calling the EloView 4 gateway (
https://secure-api.eloview.com) and not a legacy EloView 3 host. - Verify Status Codes: Isolate authorization issues (
HTTP 401) from infrastructure throttling (HTTP 429). - Review Operational Metrics: Review token reuse lifespans, individual kiosk polling frequencies, utilization of bulk endpoints, and active thread concurrency counts.
📞 Escalating to Technical Support
If unexpected API issues persist, contact Elo Technical Support. To accelerate investigation, please provide:
- API Base URL and specific endpoint called
- Your public-facing outbound source IP addresses
- Precise event timestamps (including time zone)
- Returned HTTP status codes and complete response payloads
Please report any broken links by emailing elo.support@zebra.com and include a link to the knowledge article