Open esune opened 1 year ago
Log shipping would be a good idea, so a team could have their logs streamed to an endpoint of their choosing. This could reduce the effort of having to provide an interface.
This cud be epic size work , with multiple stories linked to this item
In OCP 201 there's referenced a recommended setup that people have been using that someone created for sidecar-sending to an ELK stack
https://github.com/bcgov/elmsd-nodejs/tree/main/packages/openshift/templates/efk-stack
Other details from slide in workshop:
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.
This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.
This issue was closed because it has been stalled for 5 days with no activity.
When integrating an application/controller with a Traction tenant, it would be useful for the developers to be able to access their own agent logs in order to have a view into the agent to troubleshoot issues.
The idea would be to be able to feed on a log stream generated by the agent and present the logs relevant to each tenant using an API endpoint, and in the UI (the console-style used by OpenShift constitutes a pretty good developer experience and is what we should target for the UI).
Currently, aca-py logs are however not scoped per-tenant and there is no way of accessing them outside of the hosting platform, which constitutes a burden for operations (as it has to be involved in every step of the troubleshooting to monitor and extract the logs as actions are being executed) and a blocker for system integrators.
This functionality is a good candidate for a plug-in since it is not necessarily something that should be included in aca-py core. To be defined whether this should be traction-specific or a more general plug-in that could be added regardless of the agent's single/multi tenant mode (I am leaning towards the latter option).
c.c.: @swcurran and @WadeBarnes for thoughts and additional input