Planned support for logs & metrics of service instances in loggregator?
Dear friends of logs and metrics
At the last CF summit in Basel, SAP’s service fabric team and our team (Swisscom) had discussions about a common use case we face:
Our devs are very happy how they can access logs of their apps using cf logs or the syslog drain.
However, they do currently have no standard way to access their service instances’ logs and metrics (think cf service-logs or a syslog drain for service instances).
This is especially critical for SAP and Swisscom as service providers who produce managed services, but we assume other CF providers will also face the same use case.
Loggregator seems to be a natural fit for such extensions, however, there are certain concerns:
From an earlier quick chat with Alex (Loggregator PM) I recall that widening the scope of loggregator was problematic (especially considered the added load this would bring).
Also, we think having this decoupled from CF (application runtime) could be beneficial, so that we could reuse it for the CF container runtime and possibly other service consumers.
Are there any efforts plan to extend Loggregator in this direction or open it up for extension?