=== scuttlemonkey is now known as scuttle|afk [05:39] \quit [05:39] Heh - oops. [07:20] Hi all - is it possible for a subordinate charm to get information about the relations of its parent, and if so, are there any good recent examples of this I can look at? [12:09] hey, is it possible for a charm to be subordinate to two others, basically we want to require either this charm or that one but not both [12:13] thedac, gnuoy`: basic but functional - https://code.launchpad.net/~james-page/charms/trusty/percona-cluster/status/+merge/273825 [12:14] jamespage, fantastic, thanks [12:14] gnuoy`, gotta love jetlag ;-) [12:15] you having jetlag is certainly working out well for me! [12:16] lol [12:28] thedac, re https://code.launchpad.net/~thedac/charms/trusty/neutron-gateway/workgroup-status/+merge/273623 [12:29] commented, tl;dr drop all the status stuff related to db's - its not required and is no-op even if optionally added [12:29] we need to drop the relations but thats different work imho === JoshStrobl is now known as JoshStrobl|Nap [13:18] beisner, ping? [13:24] any idea why changing the lxc default configuration for ip range would break juju with the local provider? [13:25] I changed the bridge IP fof lxcbr0 from 10.0.2.0 to 10.0.3.0 and now charms hang at Waiting for agent initialization to finish [13:25] wait, I think I missed a config === jaywink_ is now known as jaywink === scuttle|afk is now known as scuttlemonkey [14:29] jamespage: thanks, I'll get that done. [14:37] thedac, awesome === JoshStrobl|Nap is now known as JoshStrobl [16:42] jamespage: gnuoy: For neutron-gateway making neutron-plugin-api required results in blocked even when related. It doesn't seem NeutronAPIContext is "registered" so it doesn show as a complete context. ideas? [17:09] cory_fu: tvansteenburgh: to sentry a unit in an amulet test, "self.deployment.sentry['tomcat'][0]" is preferred over "self.deployment.sentry.unit['tomcat/0']", right? [17:09] Yes [17:10] thx [17:11] The latter can fail if the charm has been deployed before in the environment, which can happen with how bundletester runs the tests and resets the environment between files within a charm or bundle test [17:16] after a reset, unit numbers *should* start over at 0, although machine numbers won't [17:17] even so, not hardcoding unit numbers is best [18:03] thedac, did you solve your neutronapi question? [18:06] gnuoy`: no I have not. Would you have time for a quick hangout? [18:22] gnuoy`: jamespage: when you have a chance https://code.launchpad.net/~thedac/charms/trusty/neutron-gateway/workgroup-status/+merge/273623 [18:50] thedac, looking [18:50] thanks [18:50] thedac, I thought that NeutronAPIContext was a base class for one of the other contexts for the charm, but it would appear not [18:55] thedac, its subtle but how about changing NeutronGatewayContext to inherit from NeutronAPIContext [18:56] api_settings = NeutronAPIContext()() [18:56] ah, ok, let me test that out. [18:56] would then become a call to its super calss [18:57] I'll run with that [18:57] thedac, super(NeutronGatewayContext self).__call__() [18:57] thedac, I think that is neater [18:57] ok === \b is now known as benonsoftware === mwenning is now known as mwenning-appt [21:00] jamespage: fyi, the neutron-gateway changes are up. [21:01] thedac, ack [21:42] Can I get a second opinion on https://code.launchpad.net/~thedac/charms/trusty/hacluster/workload-status/+merge/273889 ? [21:48] coreycb: I merged your heat MP. Looking at the keystone action managed upgrade MP now. Let me know if there are more. [21:48] thedac, thanks! [21:48] thedac, I think that's good for now [21:48] cool [21:52] gnuoy`, thedac: I commented on 2/3 midonet charms and the tintri charm - landed 1 midonet charm [21:52] thedac, looking at you proposal for neutron-gateway now [21:52] thanks [21:54] thedac, +1 but I'd like to see an amulet +1 before its landed [21:54] ack [21:55] I ran a single amulet test successfully but waiting for OSCI is a good idea here [22:27] jamespage, thedac: I pushed updates to swift-proxy [22:28] in the queue [22:36] ls [22:36] irssi should have a check for 'ls' and stop the insanity [22:48] coreycb: lint and unit_test problems with swift-proxy. [23:04] Trying again with my question from yesterday: Is it possible for a subordinate charm to get information about the relations of its parent? If so, are there any good recent examples of this? [23:05] blahdeblah: I don't *think* so, but I will defer to juju devs [23:08] blahdeblah, no that's not possible - a sub only knows about its own relations [23:08] Thanks guys - so bottom line is if I want to do that I need to add relations for both the parent and the subordinate. [23:09] ^ s/\.$/?/ [23:13] correct. You can pass information back and forth between the primary and subordinate via their relation [23:32] thedac, ok so what's left on the list? [23:33] https://code.launchpad.net/~thedac/charms/trusty/hacluster/workload-status/+merge/273889 [23:33] jamespage: Ihttps://code.launchpad.net/~gnuoy/charms/trusty/cinder-ceph/workloadstatus/+merge/273861 [23:33] thedac, ok looking at hacluster [23:34] I'd like your opinion on the approach in https://code.launchpad.net/~jjo/charms/trusty/swift-proxy/fix-multiple-devices-per-node_lp1479938/+merge/266462 [23:35] jamespage: I have to head out for a bit. I will check back in a couple of hours. [23:37] and this one https://code.launchpad.net/~gnuoy/charms/trusty/ceph-radosgw/workloadstatus/+merge/273834 [23:37] thedac, landed hacluster - looked like a good v1 ;-) [23:37] cool [23:38] ok, I'll be back a bit later