slackbot6<hsiliev> To start using Abacus you need to purge app & service usage events
slackbot6<hsiliev> Just (today) we started the installation page, so not much still. Here is a reference what we do: https://github.com/cloudfoundry-incubator/cf-abacus/blob/master/etc/concourse/scripts/cf-deploy-infra#L54-L70
slackbot6<hsiliev> Yep - billing epoch
slackbot6<hsiliev> Thanks for asking these questions. They’ll be very helpful for the new guide/page
slackbot6<hsiliev> Accessible from Abacus’ github landing page as well
cf curl -X POST /v2/app_usage_events/destructively_purge_all_and_reseed_started_apps
slackbot6<hsiliev> Added a link to billing epoch in our https://github.com/cloudfoundry-incubator/cf-abacus/wiki/Installation
slackbot8<hsiliev> Bangalore team added submit usage capability in the broker dashboard. Sofia team is still working on update of service plans in the broker and the acceptance test about it. Sofia team is doing research on: - traceability & audit capabilities for usage docs - additional days in reports windows (independent of the SLACK) We started adding documents (as part of our research tasks) in our internal design-notes repo and w
slackbot8<hsiliev> Next week I intend to finish the inception doc. I want to add details on why we think we have problems with resilience, update and scalability.
slackbot8<hsiliev> Would the first Friday after the Summit work for you and Jonathan? I intend to reuse our Friday slot
slackbot8<hsiliev> We want to have design docs. That usually means that the concepts inside won't be implemented for some time. Having them in doc might be a bit misleading
slackbot4<plamendev> <here> Hi, we noticed that collector reuse the passed oauth token to perform authentication/authorization against provisioning. Is there concrete reason for that? What do you think if we use token with system scope for internal auth/authz between collector and provisioning?
slackbot4<hsiliev> <here> Our progress this week: Abacus Removed restriction on node version: we now use node 8.9.0 speed up of 1 second in reporting ECMA 2017: can use async and await updated Docker images and Concourse pipelines Perf test improvements Removed unused modules (learning, matrix, iterable) Refactored readAllPages fixes breaker timeout on long-running operations not part of dbclient export Errors are logged i
slackbot4memory (map -> each) Replay works with no cluster enabled Work continues on: minimizing effect of load spikes (queue in front of Abacus) ZDM research Monitoring research Abacus Broker:* Improved session handling and security of Dashboard Configurable client scopes in broker Work continues on protecting provisioning (adding authorization)
slackbot4<dr.max> Also, @hsiliev I was thinking for January could we plan a demo of the Broker and UI in CAB call?