Currently all errors in registry proxy deployments (registry.redhat.io, registry.access.redhat.com, registry.connect.redhat.com, brew.registry.redhat.io) are traced in Sentry. The amount of traces we generate is exceeding Sentry's limit on how many transactions they process.

Need to lower the sampling rate for sentry and/or filter only specific errors for registry proxy that should be traced in order to decrease the amount of noise in Sentry.