[02:00] <rickyGoo> when I bootstraped a juju, what the environment is that?
[02:01] <rickyGoo> the juju status is like this:
[02:01] <rickyGoo> Model    Controller  Cloud/Region         Version  SLA default  local       localhost/localhost  2.3.4    unsupported  App  Version  Status  Scale  Charm  Store  Rev  OS  Notes  Unit  Workload  Agent  Machine  Public address  Ports  Message  Machine  State  DNS  Inst id  Series  AZ  Message
[02:01] <rickyGoo> Model    Controller  Cloud/Region         Version  SLA
[08:06] <Laney> hey, need a bit of help please :(
[08:07] <Laney> I've got a juju 1 (1.25.6) environment and I can't operate on it with juju at all
[08:07] <Laney> can SSH to the bootstrap node though but everything else just hangs
[08:08] <rick_h> Laney: is jujud running on the bootstrap node?
[08:09] <rick_h> Laney: if the jujud crashed for any reason then the client won't be able to reach it to communicate
[08:11] <Laney> rick_h: hey, yeah - and I just rebooted it to be sure
[08:18] <Laney> hmm there's quite a lot of stuff in machine-0.log
[08:18] <Laney> https://paste.ubuntu.com/p/hZQZJs7Zh3/
[08:20] <rick_h> Laney: hmm, that looks like jujud cannot contact mongodb which stores the state database.
[08:20] <rick_h> Laney: look for a juju-mongodb process and maybe there's disk space issues causing mongo to not start?
[08:20] <Laney> yes!
[08:20] <Laney> FFS kernel autoremove
[08:34] <Laney> rick_h: that worked, thanks for the pointer
[08:34] <rick_h> Laney: no problem, glad you got it working. Definitely look to getting onto 2.0 if you can.
[08:35] <Laney> yeah :/
[08:45] <TheAbsentOne> Is there some sort of documentation (besides the jujucharms docs) that formally describe the how juju works, like the internals and stuff. I'm kinda trying to describe how juju operates, how the states and hooks work for my research.
[08:47] <rick_h> TheAbsentOne: hmm, I wonder if hml is aware of anything they might use for onboarding folks and such
[08:49] <TheAbsentOne> rick_h: I would love to create a diagram to illustrate the workings. I am logging everything on a gitbook, so you guys can always review it once it's done. It might help others as well
[08:49] <rick_h> TheAbsentOne: yea, there's been a couple of things that I've seen parts of 'how juju works' but it often focuses more on the language/ideas vs the internal workings of the agents and workers
[08:50] <hml> rick_h: we do have a presentation, however it is not self standing… it requires a knowledgable human to talk to it
[08:50] <rick_h> hml: ah ok
[08:50] <TheAbsentOne> rick_h: it doesn't have to be much though, just a global idea is fine
[08:51] <TheAbsentOne> Maybe I'll just try to draw some things and let you guys fix my horrible mistakes? x)
[08:52] <rick_h> TheAbsentOne: happy to :) There's some talks in the youtube.com/jujucharms that has some intro material talks there
[08:52] <TheAbsentOne> Great! Thanks for the help
[08:56] <TheAbsentOne> Ohn one more quick question, I was watching the jujutalks about modelling charms across models but I was actually wondering how the juju gui represents this. If let's say you have a model with a mysql charm in it, and you have this other model with wordpress in it and it has a relation to the mysql charm in the other model. How is that visually represented?
[08:56] <rick_h> TheAbsentOne: the short (unfortunate) answer is that the GUI doesn't represent these yet
[08:56] <rick_h> TheAbsentOne: there's some work there to update the GUI to represent services in the model that are brought in from another
[08:58] <TheAbsentOne> rick_h: Allright that was kinda what I got from the talk as well, but thanks for verifying that!
[09:40] <morty> Hey :) I'm trying to deploy a OpenStack environment, but I'm getting this error message: " hook failed: "shared-db-relation-changed" for percona-cluster:shared-db" anyone got a clue how to fix this?
[09:40] <morty> This is my yaml-file: https://pastebin.com/z1XJRHUR
[09:45] <zeestrat> morty: I'd check out the logs for the charms that are in an error state. `juju ssh <service>/<unit>`, so for example `juju ssh keystone/0`. Then check out the relevant logs in `/var/log/juju/`.
[09:47] <zeestrat> morty: Also, you might have more luck with the gang in #openstack-charms or on the mailinglist https://docs.openstack.org/charm-guide/latest/find-us.html
[09:52] <morty> Alright, thank you zeestrat :)
[16:00] <magicaltrout> requires:
[16:00] <magicaltrout>   website:
[16:00] <magicaltrout>     interface: http
[16:00] <magicaltrout>     scope: container
[16:00] <magicaltrout> can't I connect my app to haproxy?
[17:02] <kwmonroe> hey cory_fu, magicaltrout, wanna talk about hdfs storage with me and bdx? https://hangouts.google.com/hangouts/_/canonical.com/datadirs
[17:02] <kwmonroe> totes optional
[17:42] <cory_fu> kwmonroe, bdx: Possibly of interest: https://github.com/juju-solutions/layer-storage
[17:43] <cory_fu> Probably needs some work; hasn't been touched in 2 years
[17:46] <cory_fu> kwmonroe, bdx: Another point that I just thought of, perhaps attaching storage to a running instance should be essentially a no-op until an action is run, so that the operator can indicate whether that storage should be added to the pool, have data migrated from another storage, replace another storage that has failed, etc
[17:50] <bdx> totally
[17:51] <bdx> that has a far more "managed" feel to it
[17:52] <bdx> which is probably good in this case given the possibilities of what could be
[18:10] <bdx> cory_fu: I like that approach the more I think about it
[18:10] <bdx> it takes away the dependency on the storage-attached hook
[18:11] <bdx> also gives the user control over what happens with the storage
[18:13] <bdx> and when