On 12-10-12 04:17 AM, Paul Gration wrote: > We have a few matterhorn setups that we've been testing for a while (one > at 1.3.1 and the other at a trunk checkout 1.3.x from a little after > 1.3.1 was released. What we've noticed, and this could be down to > configuration somewhere, is that when we point our epiphan capture > agents at one of the servers and it is seen by the server (added to the > capture agents list in the admin web interface), when the device is > subsequently assigned to a different server or disconnected from the > network entirely / powered off - the admin page of the server still > reports the devices as 'idle' rather than offline. > During early testing, we also had a reference capture agent PC that was > also talking to the servers but this has been the only capture agent > that has ever shown it's status as 'offline'. > Obviously during testing this has made it frustrating to see where the > capture agent is really registered and whether it's actually on / > working so we wondered what might be the cause of this?
Odd... The core keeps track of when each CA last checked in. It shouldn't matter what the CA actually is... If you can still reproduce this, check via the capture-admin/agents/$agentname.json endpoint (c.f. http://testadmin.usask.ca:8080/capture-admin/agents/testca.usask.ca.json) to see if the time-since-last-update counter keeps resetting. If it is then you have something (possibly more than one thing?) reporting itself as that CA. > Also, while I'm writing to the list, we wondered whether the server > upgrade path from 1.3.1 to 1.4 will be supported? In previous versions > this hasn't been the case. Yes, there will be an upgrade path! We know that's been a big stumbling block, so we will be addressing it with this release. G > Thanks, > > Paul >
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Matterhorn-users mailing list [email protected] http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
