[atlas] Some UDM comments
- Previous message (by thread): [atlas] Some UDM comments
- Next message (by thread): [atlas] Taking the fuzz our of client cache results
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Robert Kisteleki
robert at ripe.net
Mon May 19 15:01:36 CEST 2014
Hi, > I'm working on creating some measurements and haven't quite moved into using the API yet and wanted to provide some feedback: Thanks, it's always appreciated! > 1) When creating a UDM, when it's scheduled but not yet assigned to probe(s), it would be nice to be able to increase the # of probes. This is somewhat tricky, because the answer is: it depends. If you schedule a measurement to be started in the future, then you can always just cancel the original one and schedule a new one. One can also change the set of probes for already running measurements, Although we don't recommend it because it can be very confusing when you interpret the results. > 5) If I see a probe that is behaving poorly, eg: UDM 1665566 Probe# 3925, should I be providing feedback on this to others? I believe that such bits of information could be useful. But, since they are one-to-one messages, it's probably not so useful to do them on this list :-) Would there be a need for an in-Atlas messaging service? If so, what should it allow / prevent? We have been thinking about such a thing before, but I would not want to pre-empt a useful discussion by exposing too much of our own ideas :-) Cheers, Robert
- Previous message (by thread): [atlas] Some UDM comments
- Next message (by thread): [atlas] Taking the fuzz our of client cache results
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]