2020/06/03/#smooth-operator.txt

[04:36] <jam> morning all
[08:14] <mup> PR operator#319 opened: ops: get version from git <Created by chipaca> <https://github.com/canonical/operator/pull/319>
[08:17] <mup> Issue operator#313 closed: If ops is installed, the test run picks up the wrong one <Created by chipaca> <Closed by chipaca> <https://github.com/canonical/operator/issues/313>
[08:17] <mup> PR operator#314 closed: test_main: set up PYTHONPATH so the sub-process picks up this ops <Created by chipaca> <Merged by chipaca> <https://github.com/canonical/operator/pull/314>
[08:18] <jam> morning Chipaca
[08:18] <Chipaca> jam: moin moin :)
[08:22] <jam> Chipaca, I'd like to chat a bit about serialization of StoredState, we can start on IRC and move from there if you feel it is too involved
[08:22] <Chipaca> jam: sure
[08:23] <jam> Chipaca, so I'm pretty sure you're aware that we currently use a pickle into a sqlite database for all StoredState content
[08:23] <Chipaca> yep
[08:24] <jam> Juju will accept key=value pairs for 'state-set/get' and can take that 'state-set' as 'key=value' on the CLI or as YAML key:value from --file  (which can be '-' for stdin)
[08:24] <jam> The current spike was done by putting the pickle bytes into base64 and then writing that as the value
[08:24] <Chipaca> mhmm
[08:25] <jam> it is a pretty obtuse way to understand what your state is, if you ever wanted to do "juju run --unit u/0 state-get"
[08:25] <jam> foo: gANLCi4=
[08:26] <jam> I have some concerns about pickle support, though AIUI since we force it to be simple types with a pass by Marshal
[08:26] <jam> there shouldn't be too much trouble if you wrote a pickle with python 3.7 and read it with python 3.8
[08:26] <jam> (say you did a do-release-upgrade on the machine that was running the charm, and it is now running a different version of python)
[08:27] <Chipaca> pickle's fine with that, yes
[08:27] <jam> Chipaca, it doesn't *feel* like a 'stable format to keep your state in over time"
[08:27] <Chipaca> eh, it's stable enough (you can use 3.8 to load a pickle from 2.2 for example)
[08:27] <Chipaca> but
[08:27] <Chipaca> it's not very approachable
[08:27] <Chipaca> and that might be a bigger issue :)
[08:28] <Chipaca> jam: given that they're simple types, wouldn't yaml also work for us?
[08:28] <jam> Chipaca, so the one caveat is that yaml or json as the actual payload loses the type for set vs list
[08:29] <jam> Also, JSON keys are strictly strings IIRC, so you can't have {0: 'foo'}
[08:29] <jam> I think YAML supports non-string keys
[08:29] <Chipaca> are yaml keys also -- ah
[08:29] <Chipaca> hmm
[08:30] <Chipaca> jam: and python?
[08:30] <jam> Python has the nice behavior that json.dumps({0: 'foo'}) => '{"0": "foo"}'
[08:30] <jam> iow, it silently casts your numeric keys into string keys
[08:30] <Chipaca> no i mean, make it a python literal
[08:31] <Chipaca> and use ast.literal_eval to bring it back
[08:31] <jam> $ python3 -c "import json; print(json.dumps({'0': 'foo', 0: 'bar'}))"
[08:31] <jam> {"0": "foo", "0": "bar"}
[08:32] <jam> Chipaca, what is the cleanest way to turn a nested dict into a literal? just repr() ?
[08:32] <Chipaca> yep
[08:32] <Chipaca> bah, dicts you can str because repr is str there :) but yes, repr
[08:32] <jam> Chipaca, and... there isn't much in the way of versioning, so we have a way to change our mind about how we want to store the content.
[08:33] <Chipaca> s/have/don't have/ ?
[08:33] <Chipaca> we could add that ourselves, make the first byte of the value be the version?
[08:34] <Chipaca> does juju care at all what we dump in there? what further processing does it do on it?
[08:34] <jam> >>> repr({'0': [1,2,3], 0: set(['a', 'b', 'c'])})
[08:34] <jam> "{'0': [1, 2, 3], 0: {'b', 'c', 'a'}}"
[08:34] <jam> >>> ast.literal_eval("{'0': [1, 2, 3], 0: {'b', 'c', 'a'}}")
[08:34] <jam> {'0': [1, 2, 3], 0: {'b', 'c', 'a'}}
[08:34] <jam> Chipaca, so Juju *should* just round trip the content we give it.
[08:34] <Chipaca> yeah, sets have literals too ;)
[08:35] <jam> Chipaca, yeah, I always use set() because the empty set doesn't have another literal
[08:35] <Chipaca> ah, yeah
[08:36] <jam> since {} would be ambiguous. I'm also not a huge fan of "did you notice there were no :"
[08:36] <Chipaca> and it might fail here :-/
[08:36] <Chipaca> because 'set()' isn't a literal
[08:36] <jam> >>> ast.literal_eval(repr({'0': [1,2,3], 0: set()}))
[08:36] <jam> Traceback (most recent call last):
[08:36] <jam>   File "<stdin>", line 1, in <module>
[08:36] <jam>   File "/usr/lib/python3.6/ast.py", line 85, in literal_eval
[08:36] <jam>     return _convert(node_or_string)
[08:36] <jam>   File "/usr/lib/python3.6/ast.py", line 66, in _convert
[08:36] <jam>     in zip(node.keys, node.values))
[08:36] <jam>   File "/usr/lib/python3.6/ast.py", line 65, in <genexpr>
[08:36] <jam>     return dict((_convert(k), _convert(v)) for k, v
[08:36] <jam>   File "/usr/lib/python3.6/ast.py", line 84, in _convert
[08:36] <jam>     raise ValueError('malformed node or string: ' + repr(node))
[08:36] <jam> ValueError: malformed node or string: <_ast.Call object at 0x7f2b34d35160>
[08:36] <jam> Chipaca, indeed, it does fail :9
[08:36] <jam> :(
[08:37] <Chipaca> so close
[08:37] <jam> it is also true that we are running pickle.loads() on unknown data
[08:38] <jam> yes we write it from things that are 'safe', but we are still running the engine on data that we're reading back from 'outside'
[08:38] <jam> Chipaca, code layering wise, it is also quite hard to not use pickle, because Framework is the part that decides pickle, while Storage decides to write it to the DB
[08:39] <Chipaca> yeah
[08:39] <jam> (Framework.save_snapshot is where the marshal.dumps check and pickle.dumps is done)
[08:39] <Chipaca> i think that could use a refactor, but separately
[08:39] <jam> Chipaca, though again, we can just move that down into the SQLiteSnapshot layer
[08:39] <jam> SQLiteStorage
[08:40] <jam> Chipaca, indeed. snapshot() from the actual objects that people implement is still defined as a dict so we're ok there
[08:40] <Chipaca> jam: do we care about the spike being forwards-compatible?
[08:40] <Chipaca> i guess we do but just checking :)
[08:40] <jam> Chipaca, as in that the current version of the code can read the content written by future versions?
[08:41] <Chipaca> as in the state written by the spike be readable by future versions
[08:41] <jam> or if it can't read it, fail gracefully
[08:41] <jam> Chipaca, we don't care about the current spike, because it was just to show proof-of-concept. no charms that I'm aware of are using it.
[08:41] <Chipaca> jam: ah so the spike isn't what we'd put in 0.7
[08:42] <jam> Chipaca, correct. I'm using it as a starting point, but taking the time to carefully structure things and make sure it is the serialization that we're happy with.
[08:42] <Chipaca> ok
[08:42] <jam> I'd really rather use something like json/yaml and have it be inspectable by humans.
[08:42] <Chipaca> so for the initial spike, pickle is fine then :)
[08:42] <jam> I think you missed a comment by Stub that people are writing sqlite3 hacks to unpickle the snapshot fields to try to figure out what is going on.
[08:43] <Chipaca> i've seen that in bugs and etc, yes
[08:43] <Chipaca> which seems obvious to me -- if i were trying to figure out what was going wrong, i'd want to look at anything i could too
[08:43] <jam> Chipaca, and we can write a common helper to make it easier to inspect that state
[08:44] <jam> Chipaca, but ideally it would be less opaque to start with.
[08:44] <jam> Could we make the statement "sets aren't allowed" and just go with JSON ?
[08:44] <Chipaca> json also mangles numbers
[08:44] <jam> ah right
[08:44] <jam> yeah, I hate that part.
[08:44] <Chipaca> so it's my least favourite for this :)
[08:44] <jam> if it *failed* I might be ok with it.
[08:44] <jam> the fact that it silently corrupts makes it out
[08:45] <jam> see my above comment where you can json.dumps() and get the integer and string keys colliding with no error.
[08:45] <Chipaca> we can come up with special handling for empty sets and work with that, also
[08:45] <stub> I'd actually assumed sets were not allowed by 'simple types only', like json and yaml and toml and everything else people are used to now days.
[08:45] <davigar15> Hello!
[08:45] <davigar15> ERROR juju.worker.uniter.operation hook "start" (via explicit, bespoke hook script) failed: exit status 1
[08:45] <stub> Recommend feeding the CLI tool using the --yaml argument to avoid CLI command length limitations and encoding and locale issues
[08:45] <davigar15> I'm getting that error in k8s charms in 2.8.1
[08:46] <davigar15> Chipaca: any ideas?
[08:46] <Chipaca> davigar15: what operator version?
[08:46] <davigar15> latest
[08:46] <Chipaca> davigar15: can we see the debug logs?
[08:46] <davigar15> yes
[08:46] <jam> stub, yeah, --file is currently the way that I'm interacting with state-set.
[08:46] <jam> davi
[08:47] <jam> davigar15, I'd start with "juju model-config logging-config=DEBUG" and "juju debug-log" to get more information about what is going wrong.
[08:47] <stub> jam: Maybe this time we can use = and . in our keys :)
[08:47] <Chipaca> we _could_ restrict what you can put in state even further, for this feature
[08:47] <Chipaca> or, we could special-handle empty sets :)
[08:48] <Chipaca> (or sets in general!)
[08:48] <davigar15> oooh, good trick. Need to debug it :-)
[08:49] <Chipaca> jam: yaml.safe_load(yaml.dump(set()))
[08:49] <Chipaca> jam: → set()
[08:50] <Chipaca> yaml.safe_load(yaml.dump({'0': [1,2,3], 0: set(['a', 'b', 'c'])})) → {0: {'a', 'b', 'c'}, '0': [1, 2, 3]}
[08:51] <Chipaca> we'd have to check if the !!set annotations require a particular pyyaml version
[08:52] <davigar15> application-zookeeper-k8s: 10:52:25 DEBUG unit.zookeeper-k8s/2.start     raise TypeError(
[08:52] <davigar15> application-zookeeper-k8s: 10:52:25 DEBUG unit.zookeeper-k8s/2.start TypeError: observer methods must now be explicitly provided; please replace observe(self.on.config_changed, self) with e.g. observe(self.on.config_changed, self._on_config_changed)
[08:52] <davigar15> There is it
[08:53] <davigar15> It has to be explicit. I have self instead
[08:53] <davigar15> :-)
[08:53] <davigar15> thanks
[08:53] <Chipaca> jam: but it works in 3.11 (xenial) and 3.12 (bionic) at least
[08:53] <jam> Chipaca, ^^ that looks like the "catch exceptions and log at higher than DEBUG" :)
[08:53] <Chipaca> davigar15: :) sorry we broke your charm, glad we were trying to tell you about it at least :)
[08:54] <Chipaca> jam: yep
[08:54] <Chipaca> did we have an issue for that? i should do that one next
[08:54] <jam> we do
[08:54] <davigar15> It's okay! nothing in production yet hahaha
[08:54] <jam> Chipaca, #289
[08:54] <mup> Issue #289: hook exception logged at DEBUG <Created by jetpackdanger> <https://github.com/canonical/operator/issues/289>
[08:54] <Chipaca> perfect
[08:54] <Chipaca> jam: any reason not to go the yaml route (after the refactor to move sqlite decision out of framework) ?
[08:55] <jam> Chipaca, so I've been told historically that YAML is good if a human is going to edit it, but it has enough edge cases it doesn't make a great machine<=>machine format.
[08:55] <stub> I think you will need to keep sqlite if you want Juju 2.7 compatibility
[08:55] <jam> (things like base 60 numbers)
[08:55] <jam> stub, yes, it isn't removing sqlite
[08:56] <jam> it is adding an alternative when it is available
[08:56] <Chipaca> yeah i bungled and meant 'pickle' there
[08:56] <jam> but what do we want *that* to look like
[08:56] <Chipaca> framework is pickoing, storage is sqliting
[08:56] <Chipaca> pickling
[08:56]  * Chipaca used to be able to type
[08:56] <jam> Chipaca, I think I'll want to keep the marshal() check, but move the pickling into Storage and out of JujuStorage and try using YAML there.
[08:57] <jam> Chipaca, it kind of has to happen at the same time, since the interface to the Storage changes from taking a bytes to taking a dict
[08:57] <jam> but it shouldn't be very invasive.
[08:58] <Chipaca> jam: so what _is_ a good machine ⇔ machine format?
[08:58] <stub> With the marshal check, the yaml should always be bog simple except for sets (and maybe other builtin Python types I can't think of right now), and seems fine to me as machine <-> machine
[08:58] <jam> well, the recommendation was JSON, but for obvious reasons that is out. I don't think pickle/python ast fall into that category. XML is overbroad
[08:59] <jam> stub, Chipaca certainly YAML is my top pick right now
[08:59] <Chipaca> base64 of a pickled xml ast dump \o/
[08:59]  * Chipaca hides
[08:59] <stub> eval(repr(blob))
[08:59] <Chipaca> exec*
[08:59] <Chipaca> ]:-)
[08:59] <Chipaca> wait is exec eval now
[09:00]  * Chipaca checks
[09:00] <Chipaca> nope exec is still the right one for maximum pawnage
[09:01] <Chipaca> anyway, sorry for the silliness :)
[09:01]  * Chipaca is only a little bit sorry
[09:01] <Chipaca> hush you
[09:01] <jam> Chipaca, if only there was some way to get "sudo" in there
[09:02] <Chipaca> jam: charms run as root ¯\_(ツ)_/¯
[09:06] <jam> Chipaca, so one nice property is that refactoring is actually quite localized
[09:07] <jam> Chipaca, what is the type annotation for "snapshot". Currently we only really ever use 'dict' or None, but it has so far been anything that python can pickle
[09:08] <jam> is that just "object" ?
[09:09] <Chipaca> jam: Any ?
[09:09] <Chipaca> it's not strictly Any, but I don't think there's something better
[09:10] <Chipaca> OTOH maybe we should enforce it being dict-or-None
[09:10] <Chipaca> anything else isn't going to be extensible enough for long-term
[09:10] <davigar15> Chipaca: Why could my charm be in wating for container, without giving any errors?
[09:11] <Chipaca> 'waiting for container' sounds like a juju thing to me :)
[09:11] <Chipaca> ie not yet the charm
[09:13] <Chipaca> jam: OTOH you *could* spell it all out
[09:14] <Chipaca> as in, Union[int, float, … ]
[09:15] <jam> Chipaca, so we do have tests that snapshot() -> str and 'def snapshot(self): return self.value' where value was passed in to .emit(value)
[09:15] <jam> (where in the test value is only ever a string)
[09:18] <Chipaca> and it's probably overly opinionated for a framework :)
[09:18] <Chipaca> i mean the 'always use a dict' thing
[09:21] <jam> Chipaca, well, dict or None would be acceptable IMO, and we are little if not opinionated :)
[09:21] <jam> Chipaca, if our review of people's code would always be "you really should use a dict" then we should push that a bit harder, I think.
[09:22] <mup> PR operator#320 opened: ops/framework.py: Move pickle into Storage <Created by jameinel> <https://github.com/canonical/operator/pull/320>
[09:24] <davigar15> Chipaca: happening with juju stable version... I'm going to check my charm.
[09:26] <davigar15> Can't see anything wrong... https://pastebin.canonical.com/p/BMqcdkT6fN/
[09:29] <Chipaca> davigar15: what does the debug log say
[09:29] <davigar15> juju model-config logging-config=DEBUG
[09:29] <davigar15> ERROR config value expected '=', found "DEBUG"
[09:29] <davigar15> haha
[09:30] <jam> that exact line worked here
[09:31] <jam> you could also do: juju model-config "logging-config=<root>=DEBUG"
[09:31] <jam> which is the more explicit version
[09:31] <jam> (that allows you to specify each module separately)
[09:31] <jam> davigar15, though your particular statement "waiting for container", I believe, is created by Juju
[09:32] <jam> (if a charm sets a pod spec, and otherwise goes into Active, Juju sets the state to 'waiting for container' until the pod has started)
[09:32] <jam> davigar15, https://discourse.juju.is/t/how-to-avoid-the-waiting-for-container-message/1369/3 might help?
[09:34] <jam> specifically https://discourse.juju.is/t/how-to-avoid-the-waiting-for-container-message/1369/6
[09:35] <davigar15> https://pastebin.canonical.com/p/CxPhPygZb4/
[09:35] <davigar15> $ microk8s.kubectl -n zk get pods
[09:35] <davigar15> NAME                             READY   STATUS    RESTARTS   AGE
[09:35] <davigar15> modeloperator-658686f765-cfsr6   1/1     Running   0          2m59s
[09:35] <davigar15> zookeeper-k8s-operator-0         1/1     Running   0          2m11s
[09:35] <davigar15> But the application pods are not even created
[09:37] <jam> davigar15, so offhand your charm seems to be missing a self.state.set_default(spec=None) as on_config_changed introspects self.state.spec directly
[09:38] <davigar15> create Pod zookeeper-k8s-0 in StatefulSet zookeeper-k8s failed error: Pod "zookeeper-k8s-0" is invalid: spec.containers[0].ports[2].name: Invalid value: "leader-election-port": must be no more than 15 characters
[09:38] <davigar15> found
[09:38] <davigar15> haha
[09:38] <jam> (I don't think this the problem here, as it means hitting config_changed without start, which will happen with 2.8)
[09:39] <jam> davigar15, where did you find that? I don't see it in your pastebin
[09:55] <Chipaca> it's one of those no-rest-for-the-wicked days
[09:56] <Chipaca> taking a breka now and getting the lunch sorted because then i'll be eating in meetings
=== facundo__ is now known as facubatista
[11:01] <facubatista> Muy buenos días a todos!
[11:04] <davigar15> Chipaca: how can I get the number of ensured units with the op framework
[11:08] <jam> davigar15, is that len(self.get_relation(PEER_RELATION).units)
[11:08] <jam> or do you need the count before they are actually started?
[11:09] <davigar15> good question
[11:09] <davigar15> I think I need it before
[11:16] <jam> https://github.com/canonical/operator/issues/165 is about possibly having the framework expose the count (reading it from goal-state), but that hasn't been implemented, so understanding your use case/adding it to that issue is useful
[11:16] <jam> davigar15, ^^
[12:05] <Chipaca> facubatista: https://github.com/canonical/operator/releases/tag/0.6.0
[12:05] <facubatista> Chipaca, wonderful
[13:12] <Chipaca> facundo__: note #294 is gtg
[13:12] <mup> PR #294: Log when a served breakpoint is not really activated because of name mismatch <Created by facundobatista> <https://github.com/canonical/operator/pull/294>
[13:12] <facundo__> Chipaca, ack
[13:13]  * facundo__ will restart the router to see if all these disconnections stop
[13:13] <jam> Chipaca, and for issue #21, if it is a script, and we set JUJU_DISPATCH_PATH it will try to reinvoke hooks/install but that will notice the OPERATOR_DISPATCH env and exit early, right?
[13:13] <mup> PR #21: Persist stored state before framework commit <Created by dshcherb> <Merged by niemeyer> <https://github.com/canonical/operator/pull/21>
[13:14] <Chipaca> jam: indeed
[13:14] <Chipaca> charmcraft#21, mup m'dear
[13:14] <mup> Issue charmcraft#21: Event charm.py not defined <Created by knkski> <https://github.com/canonical/charmcraft/issues/21>
[13:14] <Chipaca> facundo__: want me to fix this ^? i can do that right now
[13:15] <facundo__> Chipaca, be my guest
[13:19] <jam> Chipaca, btw, if you want easier ways to get to source code from a git submodule you can open the '.gitmodules' file which contains the URL that git will use to find the revision
[13:19] <jam> https://git.launchpad.net/charm-k8s-mattermost/tree/.gitmodules
[13:19] <Chipaca> jam: 👍
[13:20] <jam> github does that for you, eg if you look at this: https://github.com/jameinel/ubuntu-lite/pull/2/files#diff-b1d90847e9ac9d08a6205ad8f09fd5a9L14 it has a direct link to the Operator code
[13:20] <mup> PR jameinel/ubuntu-lite#2: Operator <Created by jameinel> <https://github.com/jameinel/ubuntu-lite/pull/2>
[13:21] <jam> or https://github.com/jameinel/ubuntu-lite/tree/operator/mod if you are browsing the Tree rather than the PR
[13:23] <jam> Chipaca, while sending the mattermost email, I realized we also really need to review https://github.com/johnsca/resource-oci-image/blob/master/oci_image.py
[13:23] <jam> since that is as common to be reused as k8s.py
[13:24] <Chipaca> ah
[13:24] <Chipaca> jam: ok
=== facundo__ is now known as facubatista
[14:32] <Chipaca> closed charmcraft#18, charmcraft#19 and charmcraft#20. I hope I was able to explain why clearly enough for the reporter not to hate us too much :-|
[14:32] <mup> Issue charmcraft#18: Poetry integration <Created by knkski> <Closed by chipaca> <https://github.com/canonical/charmcraft/issues/18>
[14:32] <mup> Issue charmcraft#19: Pep 518 support for charm metadata <Created by knkski> <Closed by chipaca> <https://github.com/canonical/charmcraft/issues/19>
[14:32] <mup> Issue charmcraft#20: Support pep 517 for building charms <Created by knkski> <Closed by chipaca> <https://github.com/canonical/charmcraft/issues/20>
[14:51] <mup> PR operator#321 opened: have a Code of Conduct <Created by chipaca> <https://github.com/canonical/operator/pull/321>
[14:54] <Chipaca> facubatista, jam, reviews on #23 would mean we could ship it today
[14:55] <mup> PR #23: Add leadership to model <Created by johnsca> <Closed by johnsca> <https://github.com/canonical/operator/pull/23>
[14:55] <Chipaca> charmcraft#23
[14:55] <mup> PR charmcraft#23: build: set JUJU_DISPATCH_PATH so we get the right hook <Created by chipaca> <https://github.com/canonical/charmcraft/pull/23>
[14:55] <facubatista> Chipaca, ack
[14:56] <mup> Issue operator#220 closed: Log in WARNING when a breakpoint is not served in debug mode <Created by facundobatista> <Closed by chipaca> <https://github.com/canonical/operator/issues/220>
[14:56] <mup> PR operator#294 closed: Log when a served breakpoint is not really activated because of name mismatch <Created by facundobatista> <Merged by chipaca> <https://github.com/canonical/operator/pull/294>
[14:58] <mup> Issue operator#262 closed: Harness should support resource-get <Created by jameinel> <Closed by chipaca> <https://github.com/canonical/operator/issues/262>
[14:58] <mup> PR operator#296 closed: Oci resources <Created by chris-sanders> <Merged by chipaca> <https://github.com/canonical/operator/pull/296>
[15:08] <facubatista> Chipaca, would you mind to explain what magic are you doing in charmcraft#23?
[15:08] <mup> PR charmcraft#23: build: set JUJU_DISPATCH_PATH so we get the right hook <Created by chipaca> <https://github.com/canonical/charmcraft/pull/23>
[15:09] <Chipaca> facubatista: i see no magic :-/
[15:09] <Chipaca> facubatista: the {{}} is because I need actual {s in the output
[15:10] <Chipaca> facubatista: WRT ${parameter:-word}    Use Default Values.  If parameter is unset or null, the expansion of word is substituted; otherwise, the value of parameter is substituted.
[15:11] <Chipaca> so ${JUJU_DISPATCH_PATH:-$0} is "either JUJU_DISPATCH_PATH or, if that is unset or null, $0"
[15:13] <facubatista> Chipaca, fantastic, thanks!!!
[15:14] <facubatista> Chipaca, mind to put a comment saying something like "if Juju don't support the dispatch mechanism, it will execute the hook, and we need sys.argv[0] to be the name of the hook, so we fake JUJU_DISPATCH_PATH to be that value"?
[15:14] <facubatista> Chipaca, thanks!
[15:15] <Chipaca> will do
[15:16] <facubatista> Chipaca, approved
[15:16]  * facubatista -> lunch
[15:58] <Chipaca> hmmmmmm
[15:58]  * facubatista is back
[15:58]  * Chipaca hmms
[16:01] <Chipaca> dunno if my lxd is stuck, or just my juju
[16:01] <Chipaca> facubatista: do you have a working thing you could juju ssh into to try something with juju-log?
[16:02] <facubatista> Chipaca, yes, but on a meeting right now
[16:02] <Chipaca> ah
[16:02] <Chipaca> k
[16:18]  * facubatista brb
[17:02] <Chipaca> ok i think i'm off
[17:02]  * Chipaca EODs
[17:03]  * facubatista is back
[19:44]  * facubatista eods