Eduroam development VC, 20170711
Attendees
--------------
Stefan Winter, RESTENA
Reimer Karlsen-Masur, DFN-CERT Services GmbH
Miroslav Milinovic, Srce / CARNet
Jørn Åne, UNINETT
Maja Gorecka-Wolniewicz, PIONIER
Brian Epstein, Institute for Advanced Study
Marko Eremija, AMRES
Pedro Simões, FCCN
Žilvinas Vaira, LITNET
Mike Zawacki, Internet2
Zenon Mousmoulas, GRNET
Tomasz Wolniewicz, PSNC
Apologies
--------------
Agenda / Minutes
----------
1. Welcome, agenda bashing
2. Managed IdP: SMS sending implemented
no insight into current balance at Nexmo - cat-pilot could be "broke" at some point. Stefan to top up balance and think about an alerting mechanism
1:n mapping of invitation vs. cert implemented in backend - needs UI (especially: space for number of uses; deletion of partly used invitation)
lets make revocation of certificates and revocation of invitations independent. If a admin wants to "kill" a user he should remember to revoke both the certificates AND outstanding open invitation links
"Deactivate user" should however revoke both certificates AND the pending invitations
3. Self-service diagnosis
- probe data format
JSON is fine for techs, but human-readable documentation should accompany it
need to differentiate between: dedicate hardware devices - probes - vs. end-user devices doing proby stuff: we are right now only working on the former; latter is a "GN6 dream" :-)
Zenon mentions statsd; probably enough if it is just about metrics (boolean, integer, ...)
https://github.com/etsy/statsd ? https://www.digitalocean.com/community/tutorials/how-to-configure-statsd-to-collect-arbitrary-stats-for-graphite-on-ubuntu-14-04 ?
Defining "metrics":
http://metrics20.org/spec/
However metric types vary per implementation:
https://github.com/etsy/statsd/blob/master/docs/metric_types.md
https://prometheus.io/docs/concepts/metric_types/
https://help.datadoghq.com/hc/en-us/articles/206955236-Metric-types-in-Datadog
These have native boolean types:
https://www.elastic.co/guide/en/elasticsearch/reference/current/boolean.html
https://cloud.google.com/monitoring/api/v3/metrics
Zenon's point was: If you only need to collect some metrics (counters, gauges... *not* arbitrary strings) use some widely implemented/available metric format/protocol; if you need to process *structured data*, then perhaps you need another schema.
4. AOB / Next VC
* 25 July 2017, 1530 CEST
