HaloPSA limits how many API calls an integration can make against a tenant in a short window. CloudRadial's user and ticket sync counts against that limit. This article explains what the limit is, when a CloudRadial sync is likely to hit it, what you will see, and what to do.
In this article:
- What the limit is
- When CloudRadial sync hits it
- What you will see
- What to do
- Frequently asked questions
What the limit is
Halo enforces a limit of 700 API requests per rolling 5 minute (300 second) window per tenant. When the limit is exceeded, Halo responds with HTTP 429 (Too Many Requests) and rejects further calls until the window clears. This is a Halo platform limit documented in Halo's API reference, not a CloudRadial setting, and it cannot be raised from the CloudRadial side.
The limit is shared by every integration using your Halo API application. CloudRadial's calls, any RMM or billing integrations you run through the same API key, and your own scripts all draw from the same 700.
When CloudRadial sync hits it
Normal scheduled syncs on an established client rarely approach the limit. The situations that do:
- Initial sync of a large client. A client with several hundred contacts, plus their tickets and other PSA records, can require more than 700 calls in one pass.
- Repeated manual syncs. Triggering sync on a client several times inside a few minutes, for example while testing a fix, uses the window up quickly.
- Several large clients syncing at once. Multiple companies syncing in the same window share the same tenant limit.
- Other integrations on the same API key. If another tool is already consuming most of the 700, CloudRadial's share is smaller.
What you will see
- A manual sync that starts but ends in an error, or appears to finish with far fewer users or tickets than expected.
- A user or ticket that exists in Halo and should match the client's settings but does not appear in CloudRadial after a sync.
- Sync log entries ending in an error from the PSA during the sync run. The sync log is available from the client's 3-dot menu under Partner > Clients.
The failure can look like a matching or configuration problem. If the client's sync settings and PSA identifiers are correct and the sync still fails during a large or repeated run, suspect the rate limit before changing settings.
What to do
- Wait at least 5 minutes before re-running. The window is rolling, so the limit clears on its own. Re-triggering sync immediately only pushes the window further out.
- Do not stack manual syncs. Run one sync, wait for it to finish, then review the result before running another.
- Check what else is using the API key. In Halo, go to Configuration > Integrations > HaloPSA API and review the applications. If other tools share the application CloudRadial uses, consider giving CloudRadial its own API application so its calls are isolated.
- Let the overnight sync do large first loads. For a new client with a large contact or ticket history, enabling sync and allowing the scheduled sync to run avoids competing with daytime API traffic.
- Contact CloudRadial support if a large initial sync keeps failing.
Frequently asked questions
Q: Can CloudRadial raise the limit?
No. The limit is enforced by Halo per tenant. Halo's API documentation covers it under Rate Limit.
Q: Will I lose data when a sync is rate limited?
No. Nothing is deleted. Records that were not retrieved in the failed run are picked up by the next successful sync.
Q: Does this affect ticket submission from the portal?
Ticket creation uses only a handful of calls per ticket, so day-to-day portal use is not normally affected. A rate limit reached by a large sync can briefly delay ticket actions until the window clears.
Q: Does this apply to HaloPSA on-premise?
Self-hosted Halo instances enforce limits based on their own configuration. If you host Halo yourself, check your instance settings or ask Halo support.
Comments
0 comments
Article is closed for comments.