/srv/irclogs.ubuntu.com/2019/11/07/#juju.txt

thumperbabbageclunk: did the full application databag stuff get into the 2.7 rc?00:50
thumperwallyworld: ^^00:50
wallyworldthumper: otp, but yes, except for cross model00:50
babbageclunkthumper: I'm doing the CMR part now.00:50
thumperack00:50
* thumper is going to mark it complete00:51
anastasiamacthumper: can u ho?00:52
thumpernot just now00:52
wallyworldhpidcock: kelvinliu: jump into standup?01:03
kelvinliuyep01:03
timClicksnew tutorial up - https://discourse.jujucharms.com/t/deploy-a-redis-cluster-at-any-scale-on-any-cloud/232002:30
timClicksappreciate anyone's time to check for typos and also to make sure that the video plays properly02:34
hpidcockwallyworld: kelvinliu: did we want to continue the discussion?02:47
kelvinliuyeah, im free02:48
hpidcockI guess we can do it tomorrow. I've put down as much as I can down in the doc.03:01
wallyworldhpidcock: sorry, been tied up04:16
hpidcockall good, I got everything down I wanted to say04:51
hpidcockwallyworld: I'm considering adding a Hash field to juju/version's Number, because its passed around almost everywhere and would make this incredibly easy. But I'm also very weary that version.Number is used in so many places and would also make comparing weird because Hash isn't monotonic.04:57
wallyworldyeah that04:58
wallyworldjam: i had one thing to discuss which i only just recalled, but you are probs busy now?04:59
hpidcockyeah that?04:59
wallyworldthe last bit about comparisons04:59
wallyworldwe compare version numbers04:59
hpidcockbut Version also has Tags04:59
jamwallyworld: sure, I'll jump back in05:00
hpidcockalpha, beta, rc are just... conveniently monotonic haha05:00
wallyworldyeah, i'm sure we've used that before :-)05:05
manadartstickupkid: Review? https://github.com/juju/juju/pull/1087709:48
stickupkidmanadart, sure09:49
stickupkidmanadart, some questions09:53
nammn_demanadart: what is an "in-flight" branch again? Is it a not commited/aborted branch?10:33
manadartnammn_de: Yes.10:33
nammn_demanadart: how do I access all branches, not only in-flight ones? I cannot seem to find the proper implementation from apiserver modelgernation10:35
nammn_de This is the place I was initially trying: https://github.com/nammn/juju/blob/e012d8047bd6b8594fabdeeebf319c76ece98834/apiserver/facades/client/modelgeneration/modelgeneration.go#L293 .10:36
manadartnammn_de: I think you will need a new state method to retrieve committed ones.10:39
nammn_demanadart: got it10:39
nammn_deI was so confused cause my ide (intellij) kept showing me that we only implemented branches interface  as a mock. Just couldnt find the impl :D10:40
achilleasamanadart: quick question. Is Subnet.SpaceTag (in apiserver/params/network.go) pointing to the space ID?10:40
manadartnammn_de: It fails like that sometimes. I've had the same thing.10:41
manadartachilleasa: SpaceTag still implies a name.10:41
achilleasagot 5 min for a quick ho?10:42
achilleasanvm I think I found what I needed10:54
manadartjam: In a transaction are "a": "d+" and "a": "d-" assertions that the docs do and do not exist respectively?11:10
nammn_demanadart: should I name all the new things "branch" or still "generations"?12:10
manadartnammn_de: For now just keep the convention - branch refers to those not yet committed. Leave the rest as-is.12:11
jammanadart: correct. those are special cased in the txn layer12:55
jammanadart: I believe there constants DocExists to be clearer about it12:56
achilleasajam: manadart is it possible to have CIDR overlaps when carving spaces? e.g. 10.0.0.0/16 -> foo but 10.0.0.42.0/24 -> bar13:25
jamachilleasa: are you talking actual networking, or logically?13:25
jamactual networking the subnets must not overlap13:25
jamotherwise it wouldn't know how to route traffic13:25
achilleasaso, if I am trying to figure out the space from the address (assuming we don't have different subnets with the same CIDR), can I use CIDR comparisons to find the space and keep the first match?13:27
achilleasa(if overlaps where possible I would need to find the CIDR with the smallest range)13:28
N3tw0rKanyone know of a charm for the ceph nautilus manager gui plugin?15:17
hmlquick review please… https://github.com/juju/juju/pull/10880  just want to get 10872 into develop.16:15
achilleasahml: shouldn't that be a forward port? :D16:20
hmlachilleasa: :-)16:21
rick_hN3tw0rK:  not sure. Might hit up discourse or the openstack charming community.16:29
hmlstickupkid: pls let me know if you’re availalble for debug help on test_copywrite16:46
stickupkidhml, i'm around16:47
hmlstickupkid: it’s failing locally for me, and i have no idea why.  i’d be more than happy to fix it or my machine if i could get that far.  :-)  https://pastebin.canonical.com/p/v4W67YTB99/16:47
hmland it’s on ubuntu not macOS.  :-D16:48
stickupkidrun this16:48
stickupkidfind . -name '*.go' | grep -v -E "(./vendor|./acceptancetests|./provider/azure/internal|./cloudconfig)" | sort | xargs grep -L -E '// (Copyright|Code generated)'16:48
hmlstickupkid: it just complete without seen output16:49
hmlstickupkid: that’s what confused me16:50
stickupkidfind . -name '*.go' | grep -v -E "(./vendor|./acceptancetests|./provider/azure/internal|./cloudconfig)" | sort | xargs grep -L -E '// (Copyright|Code generated)' | wc -w16:53
stickupkidwhat's the output of that?16:53
hml016:54
stickupkidyay16:54
stickupkidnot16:54
achilleasamanadart: is there any helper in state for getting a SpaceInfos instance? (the list; without having to go via SpaceInfosBy...)16:54
hmlha16:54
stickupkidhml https://pastebin.canonical.com/p/2GpsGwWBGk/16:55
achilleasamanadart: ah... AllSpaces16:57
hmlstickupkid: same result, not sure it’s that, i’m not getting the output at 12/13 in diff16:58
stickupkidhml, HO?17:03
hmlstickupkid: sure17:03
nammn_destickupkid after bumping the version, how do i make sure that my client can use it?17:03
nammn_debumping rev17:03
stickupkidhttps://pastebin.canonical.com/p/gHq3pfC7VJ/17:20
stickupkidrick_h, can you cross model migrate a SAAS, i'm guessing you can right?17:21
rick_hstickupkid:  yes, should be able to17:30
stickupkidrick_h, ok, so i need to fix that then :D17:30
rick_hstickupkid:  yep, that's the card17:30
rick_hthe work to make that happen wasn't done so we're closing that gap17:30
hmlstickupkid:  approved17:40

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!