d365-limits-performance
/d365-limits-performance
Keep Dataverse / Dynamics 365 integrations, migrations and queries inside the platform limits - service protection API limits (429 Too Many Requests, Retry-After, 6,000 requests / 20 minutes execution / 52 concurrent per user per 5 minutes per web server), throttling-aware retry, parallelism versus batching, bulk messages, and query anti-patterns that cause timeouts (leading wildcards, too many columns, sorting by choice or joined columns, filtering on calculated columns). Use this skill whenever the user builds a data load, migration, nightly sync, ETL into or out of Dataverse, sees 429 / "Nu
- Author
- antonioddiniz ↗
- Repository
- d365-skills ↗
- Licence
- MIT
- Last repository change
- 30.10.2026
Assessment and security
- Files in this package
- 2
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 7
No signals detected. Repository changed on 30.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Keep Dataverse / Dynamics 365 integrations, migrations and queries inside the platform limits - service protection API limits (429 Too Many Requests, Retry-After, 6,000 requests / 20 minutes execution / 52 concurrent per user per 5 minutes per web server), throttling-aware retry, parallelism versus batching, bulk messages, and query anti-patterns that cause timeouts (leading wildcards, too many columns, sorting by choice or joined columns, filtering on calculated columns). Use this skill whenever the user builds a data load, migration, nightly sync, ETL into or out of Dataverse, sees 429 / "Nu
Original author description
Keep Dataverse / Dynamics 365 integrations, migrations and queries inside the platform limits - service protection API limits (429 Too Many Requests, Retry-After, 6,000 requests / 20 minutes execution / 52 concurrent per user per 5 minutes per web server), throttling-aware retry, parallelism versus batching, bulk messages, and query anti-patterns that cause timeouts (leading wildcards, too many columns, sorting by choice or joined columns, filtering on calculated columns). Use this skill whenever the user builds a data load, migration, nightly sync, ETL into or out of Dataverse, sees 429 / "Nu
How it works
- The limits (defaults, per user, per web server, 5-minute sliding window)
- How to move a lot of data
- Query anti-patterns (Dataverse names them in the error)
- Review checklist for a job or integration
- Output
- References in this skill
- Sources (Microsoft Learn)
SKILL.md preview
--- name: d365-limits-performance description: Keep Dataverse / Dynamics 365 integrations, migrations and queries inside the platform limits - service protection API limits (429 Too Many Requests, Retry-After, 6,000 requests / 20 minutes execution / 52 concurrent per user per 5 minutes per web server), throttling-aware retry, parallelism versus batching, bulk messages, and query anti-patterns that cause timeouts (leading wildcards, too many columns, sorting by choice or joined columns, filtering on calculated columns). Use this skill whenever the user builds a data load, migration, nightly sync, ETL into or out of Dataverse, sees 429 / "Number of concurrent requests exceeded" / "Combined execution time" / PerformanceValidationIssuesCauseTimeout / LeadingWildcardCauseTimeout errors, or asks why a Dataverse query or job is slow. --- # Dataverse limits and performance The platform tells you how fast you may go. The client that listens to `Retry-After` finishes first. The client that hammers gets longer and longer penalties. ## The limits (defaults, per user, per web server, 5-minute sliding window) | Measure | Default | Error | |---|---|---| | Requests | 6,000 | 429, Retry-After | | Combined execution time | 20 minutes (1,200 s) | `-2147015903` | | Concurrent requests | 52 or higher | `-2147015898`, returned immediately | These numbers can change and vary by environment. Do not hard-code them as a rate limiter. Plug-ins and workflows calling Dataverse internally are not subject to these limits. Entitlement (licensing) limits are separate and batching does not avoid them. ## How to move a lot of data 1. **Let the server set the pace.** Start low, ramp up, and on 429 wait exactly `Retry-After` seconds before sending more from that user. Never retry immediately in a tight loop. 2. **Parallel single requests beat big batches.** Bound the degree of parallelism with the value of the `x-ms-dop-hint` response header, which is the environment's recommended parallelism. `ServiceClient` (and `CrmServiceClient`) already wait `Retry-After` and resend for you. 3. **Batches: start at ~10 operations,** not 1,000. Large batches hit the execution-time limit instead of the request limit. 4. **Prefer bulk messages** (`CreateMultiple`, `UpdateMultiple`, `UpsertMultiple`) on tables that support them. Data must be clean: one error fails the call. 5. **Synchronous plug-ins are your hidden cost.** During a load, each row pays for every sync step. Consider the bypass optional parameter (privileged users only) or disabling steps during the window, and say clearly what logic is skipped. 6. **Several application users multiply throughput only if the business accepts it.** Limits are per user. Do not create identities just to evade governance. ## Query anti-patterns (Dataverse names them in the error) | Anti-pattern id | Fix | |---|---| | `PerformanceLeadingWildCard` | No `like '%x'`, `ends-with`, or values starting with `_`, `[`, `-%`, `'%`. Use Dataverse search or redesign the data. | | `LargeAmountOfAttributes` | Select only needed columns. Never `all-attributes` or no `$select`. | | `LargeAmountOfLogicalAttributes` | Fewer logical columns (values stored in other tables). | | `PerformanceLargeColumnSearch` | Do not filter on large text (memo) columns. | | `OrderOnEnumAttribute` | Do not sort by choice columns. | | `OrderOnPropertiesFromJoinedTables` | Do not sort by columns from link-entity / `$expand`. | | `FilteringOnCalculatedColumns` | Filter on stored columns, not calculated or formula columns. | Also: filter on indexed columns (primary keys, lookups, alternate keys), limit join depth, and page with cookies. Simple FetchXML paging stops at 50,000 rows. ## Review checklist for a job or integration - [ ] 429 handling reads `Retry-After` (Web API header or `OrganizationServiceFault.ErrorDetails["Retry-After"]`)? - [ ] Bounded parallelism, no unbounded `Task.WhenAll` over thousands of rows? - [ ] Batch size small and measured, not 1,000 by default? - [ ] Bulk messages used where the table supports them? - [ ] Every query has `$select`/attributes, deterministic order, cookie paging? - [ ] None of the seven anti-patterns above? - [ ] Sync plug-ins on the target table counted in the load plan? ## Output For a job design: throughput plan (parallelism, batch size, bulk or not), retry policy, the list of sync logic that runs per row, and the query fixes. For an error: the limit hit, the cause, the fix. ## References in this skill - `references/retry-patterns.md` - retry loops for .NET SDK, Web API and PowerShell. ## Sources (Microsoft Learn) - Service protection API limits: https://learn.microsoft.com/power-apps/developer/data-platform/api-limits - API limits overview (service protection vs entitlement): https://learn.microsoft.com/power-apps/maker/data-platform/api-limits-overview - Send parallel requests: https://learn.microsoft.com/power-apps/developer/data-platform/send-parallel-requests - Optimize performance for bulk operations: https://learn.microsoft.com/power-apps/developer/data-platform/optimize-performance-create-update - Query anti-patterns: https://learn.microsoft.com/power-apps/developer/data-platform/query-antipatterns - Query throttling: https://learn.microsoft.com/power-apps/developer/data-platform/query-throttling - Bypass custom business logic: https://learn.microsoft.com/power-apps/developer/data-platform/bypass-custom-business-logic
Files in this package (2)
- SKILL.md 5 KB
- references/retry-patterns.md 2 KB
Versions
The database has no other skills named d365-limits-performance.
Database snapshot: 13.11.2026 · package version fb5da7a1d732 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 96dataverse-python-performance-optimization
dataverse-python-performance-optimization guidelines Triggers on: ** - 91add-dataverse
Use when the user wants to add Dataverse tables (existing or new) to a Power Apps mobile app, extend an existing Dataverse table with new columns, or apply an approved data model plan. - 90d365-flow-reliability
Make Power Automate cloud flows that touch Dataverse / Dynamics 365 reliable and cheap to run - trigger filtering (select columns, filter rows, trigger conditions, scope), try/catch/finally with Scope and Configure run after, retry policies, idempotency against duplicate… - 96dataverse-python-modules
dataverse-python-modules guidelines Triggers on: ** - 100microsoft-dynamics-webhooks
Receive and authenticate Microsoft Dynamics 365 (Dataverse) webhooks registered with the Plug-in Registration Tool. Use when building a Dynamics 365 or Dataverse webhook receiver, securing it with the HttpHeader authentication option (Dataverse has no HMAC signature), parsing… - 100dataverse-overview
Background orientation for Microsoft Dataverse work: the dataverse skill map, hard rules, tool-capability matrix (MCP, Python SDK, dataverse CLI, PAC CLI, Azure CLI, GitHub CLI, Web API), SDK method cheat-sheet, and the safe change lifecycle. Load first for any Dataverse task,…