Can anyone clarify the expected timeout behavior for the Update Record data action targeting Salesforce? The integration is configured with a 30-second timeout, yet the Architect flow logs indicate a 504 Gateway Timeout error after exactly 60 seconds.
This discrepancy suggests the platform is ignoring the configured timeout value or applying a global override. The flow handles high-volume lead routing, and these delays are causing queue metrics to skew significantly in the Performance Dashboard.
The environment is running the latest release in the EU-West region. We need to determine if this is a known limitation of the Salesforce connector or a configuration error within the data action definition.
“timeout_seconds”: 60
Ah, this is a known issue. The Salesforce adapter enforces a hard 60-second ceiling regardless of the data action configuration, so any payload exceeding that threshold will trigger a 504.
this looks like a platform constraint rather than a config error. the 60s hard limit is enforced by the underlying integration runtime. try switching to a pure http request action with a custom timeout field. it bypasses the data action wrapper and respects your 30s setting, assuming the salesforce endpoint responds in time.
{
"cache_ttl": 5400,
"hard_limit_override": false
}
The Salesforce adapter definitely enforces that 60-second ceiling, just as mentioned above. In Zendesk, custom triggers didn’t have such rigid platform ceilings, so this feels like a steep learning curve during migration.
Switching to a direct HTTP action is a solid workaround. It bypasses the data action wrapper entirely, letting you control the timeout precisely. Just ensure your Salesforce endpoint handles the request efficiently to avoid those 504 errors.