Description
We’re currently encountering a rare issue in production (1–2% occurrence rate) that we are unable to consistently reproduce in development or staging environments. To help us trace and report these cases more effectively, we’re using reportToGlobalTracing to send telemetry logs to Zoom whenever our users encounter unexpected behavior.
My question is: Are there any concerns or limitations regarding the volume or frequency of logs sent via reportToGlobalTracing?
For example, if we scale this to thousands of users and only 1–2% experience the issue, we may end up sending a non-trivial number of logs.
We’re hoping to rely on these logs to collaborate with Zoom support when issues are reported by our users.
Hi @vic.yang would you be able to help clarify this? I’m planning to proceed with a release soon and want to make sure using reportToGlobalTracing at this scale won’t cause any issues. Thanks in advance!
Are there any concerns or limitations regarding the volume or frequency of logs sent via reportToGlobalTracing?
We don’t have any limitations on the frequency or number of logs sent, so there’s no need to worry about that. Zoom has sufficient capacity to handle these logs.
You mentioned, “We highly recommend sending the logs either after the issue occurs or before leaving the session.”
If we send logs frequently—every time an issue occurs, or wait to before leaving the session—could the detailed logs become too large and cause bandwidth issues?
Do we need to set telemetry_tracking_id in order to send detailed logs?
The telemetry_tracking_id field is optional. If it is not set, the Video SDK will use the session ID by default.
If we send logs frequently—every time an issue occurs, or wait to before leaving the session—could the detailed logs become too large and cause bandwidth issues?
The log content is not large, so there’s no need to worry about bandwidth issues.