| *** nkinder has joined #openstack-dns | 00:13 | |
| *** iraw_ has quit IRC | 00:25 | |
| *** MentalRay has joined #openstack-dns | 00:43 | |
| *** iraw_ has joined #openstack-dns | 01:26 | |
| *** stanzgy has joined #openstack-dns | 01:40 | |
| *** stanzgy_ has joined #openstack-dns | 01:40 | |
| *** MentalRay has quit IRC | 01:50 | |
| *** stanzgy_ has quit IRC | 02:30 | |
| *** stanzgy has quit IRC | 02:30 | |
| *** stanzgy has joined #openstack-dns | 02:31 | |
| *** andrewbogott_afk is now known as andrewbogott | 02:50 | |
| *** iraw_ has quit IRC | 02:51 | |
| *** andrewbogott is now known as andrewbogott_afk | 02:54 | |
| *** andrewbogott_afk is now known as andrewbogott | 03:01 | |
| *** andrewbogott is now known as andrewbogott_afk | 03:02 | |
| *** richm1 has quit IRC | 04:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/designate: Imported Translations from Transifex https://review.openstack.org/166506 | 06:04 |
|---|---|---|
| *** stanzgy has quit IRC | 06:46 | |
| *** stanzgy has joined #openstack-dns | 06:47 | |
| *** ttrumm has joined #openstack-dns | 07:34 | |
| *** chlong has quit IRC | 07:58 | |
| *** jordanP has joined #openstack-dns | 08:21 | |
| *** jordanP has quit IRC | 09:27 | |
| *** jordanP has joined #openstack-dns | 09:28 | |
| *** kbyrne has joined #openstack-dns | 09:35 | |
| *** iraw_ has joined #openstack-dns | 10:10 | |
| *** stanzgy has quit IRC | 10:49 | |
| *** ttrumm has quit IRC | 10:53 | |
| *** untriaged-bot has joined #openstack-dns | 11:02 | |
| untriaged-bot | Untriaged bugs so far: | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1434189 | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1434479 | 11:02 |
| openstack | Launchpad bug 1434189 in Designate "Delete zone process can be interrupted" [Undecided,In progress] - Assigned to Tim Simmons (tim-simmons-t) | 11:02 |
| openstack | Launchpad bug 1434479 in Designate "designate stop working after insert ~10k records" [Critical,New] | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432842 | 11:02 |
| openstack | Launchpad bug 1432842 in Designate "Property setters dont work." [Undecided,New] | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432850 | 11:02 |
| openstack | Launchpad bug 1432850 in Designate "Add periodic refresh task" [High,New] - Assigned to Endre Karlson (endre-karlson) | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1425668 | 11:02 |
| openstack | Launchpad bug 1425668 in Designate "Poor error message when using same database for designate and the pool manager cache" [Undecided,New] | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432839 | 11:02 |
| openstack | Launchpad bug 1432839 in Designate "Calling find_domain in any v1 operation is sub-optimal" [Medium,New] - Assigned to Endre Karlson (endre-karlson) | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1433645 | 11:02 |
| openstack | Launchpad bug 1433645 in Designate "Floating IP Reverse DNS API does not respect async operations" [Undecided,New] | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1430976 | 11:02 |
| openstack | Launchpad bug 1430976 in Designate "Pool Manager creates/deletes can go into an infinite loop" [Undecided,Fix committed] - Assigned to Tim Simmons (tim-simmons-t) | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1425117 | 11:02 |
| openstack | Launchpad bug 1425117 in Designate "Designate does not work with postgres" [High,New] | 11:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1433869 | 11:02 |
| openstack | Launchpad bug 1433869 in Designate "too many "Including all tenants items" printed in logs" [Undecided,In progress] - Assigned to stanzgy (stanzgy) | 11:02 |
| *** untriaged-bot has quit IRC | 11:02 | |
| *** mwagner_lap has quit IRC | 11:30 | |
| *** EricGonczer_ has joined #openstack-dns | 11:54 | |
| *** EricGonczer_ has quit IRC | 12:02 | |
| *** richm has joined #openstack-dns | 12:07 | |
| *** iraw_ has quit IRC | 12:19 | |
| *** EricGonczer_ has joined #openstack-dns | 12:22 | |
| *** mwagner_lap has joined #openstack-dns | 12:23 | |
| *** EricGonczer_ has quit IRC | 12:24 | |
| mugsie | andrewbogott_afk: ping me when you are around :) | 12:29 |
| mugsie | i think it is zone_records / domain_records (depending on what API version you are using) | 12:30 |
| mugsie | dvorak: not in the V1 API :( | 12:33 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Fix handling of wildcards in api / storage https://review.openstack.org/166806 | 12:54 |
| dvorak | mugsie: thanks, I think I've figured out a way around it for now | 13:00 |
| openstackgerrit | Merged openstack/designate: Spelling & formatting doc fixes https://review.openstack.org/166571 | 13:00 |
| openstackgerrit | Merged openstack/designate: Updated from global requirements https://review.openstack.org/166430 | 13:02 |
| *** jschwarz has joined #openstack-dns | 13:27 | |
| openstackgerrit | Merged openstack/python-designateclient: Updated from global requirements https://review.openstack.org/166466 | 13:37 |
| openstackgerrit | Merged openstack/designate: Remove "Including all tenants items" debug log https://review.openstack.org/165679 | 13:37 |
| dvorak | mugsie: I was thinking about it later, and that's really how it works for most other services. If I want to create a neutron network owned by a specific tenant, I have to be in that tenant. | 13:42 |
| mugsie | dvorak: we do allow for domains to be created, and then trnasfered to an other tenant in the V2 API | 13:42 |
| mugsie | and we did talk about allowing admin tennants to "sudo" to another tenant id | 13:43 |
| dvorak | is there any client library support for the V2 API yet? I looked the other day and didn't have any real luck | 13:43 |
| mugsie | not yet | 13:43 |
| dvorak | I was wondering about that, I saw sudo in the policy.json but didn't find anything in the source about it | 13:43 |
| mugsie | it is still under going massive change | 13:43 |
| dvorak | but I was looking at stable/juno | 13:43 |
| ekarlso | mugsie: s/not yet/there is a review/g | 13:43 |
| ekarlso | but nothing stable no :p | 13:43 |
| mugsie | I am about to push up a massive change ;) | 13:44 |
| mugsie | changes the v2 API quite a lot | 13:44 |
| ekarlso | mugsie: the api v2 change for the {"zone": {}} thing ? | 13:44 |
| mugsie | yeah | 13:44 |
| mugsie | and removing views | 13:44 |
| ekarlso | mugsie: should be easy to fix in my bindings though :p | 13:44 |
| mugsie | ekarlso: I will take your word for it :) | 13:44 |
| ekarlso | mugsie: just to fixup the use of "reponse_key" and the other stuff :p | 13:44 |
| ekarlso | mugsie: ping me when you got it up :D | 13:45 |
| mugsie | will do | 13:45 |
| ekarlso | mugsie: challenge accepted :p | 13:45 |
| mugsie | just trying to make it a bit smaller - its currently +90,-1500 | 13:45 |
| mugsie | +900* | 13:45 |
| *** GonZo2K has quit IRC | 13:50 | |
| *** vinod1 has joined #openstack-dns | 13:56 | |
| *** EricGonczer_ has joined #openstack-dns | 14:04 | |
| *** paul_glass has joined #openstack-dns | 14:19 | |
| *** andrewbogott_afk is now known as andrewbogott | 14:20 | |
| openstackgerrit | Tim Simmons proposed openstack/designate: Omit deleting zones from mass ns record updates https://review.openstack.org/165934 | 14:22 |
| *** kodoku has joined #openstack-dns | 14:25 | |
| timsim | So looks like there are four Designate talks at the upcoming summit: https://www.openstack.org/summit/vancouver-2015/schedule/?awesm=awe.sm_eLi9J | 14:26 |
| timsim | Unfortunately the workshop is scheduled at the same time as the talk with Kiall, rjrjr, and myself. Can we get that changed? | 14:27 |
| vinod1 | Clayton's talk is also at the same time as Ron's other talk | 14:30 |
| vinod1 | So we have 2 time slots for 4 talks | 14:30 |
| timsim | Derp | 14:30 |
| mugsie | timsim: yeah, we can usually get them moved around | 14:44 |
| mugsie | Beth seems to email a few days later, and she can usually get them moved | 14:44 |
| timsim | Good stuff. | 14:45 |
| mugsie | I have gotten them moved for the last 2 summits anyway | 14:45 |
| andrewbogott | mugsie: thanks for the response! I don’t see ‘quota-zone-records’ anyplace in the source… running version Icehouse. | 14:52 |
| mugsie | hummm | 14:52 |
| andrewbogott | I’ll try it anyway :) | 14:53 |
| mugsie | oh, in icchouse - i think it is quota-domain-records | 14:54 |
| andrewbogott | oh, dammit, I have a typo in my config file. So, let me try fixing that first :) | 14:56 |
| andrewbogott | mugsie: ok, for good measure I now have | 15:00 |
| andrewbogott | quota-domain-records = 4000 | 15:00 |
| andrewbogott | quota-domain-recordsets = 4000 | 15:00 |
| andrewbogott | quota-zone-records = 4000 | 15:00 |
| andrewbogott | quota-zone-recordsets = 4000 | 15:00 |
| andrewbogott | quota-domains = 1000 | 15:00 |
| andrewbogott | But it still won’t go over 500 | 15:00 |
| andrewbogott | tbh I’m not clear on how it manages this per-tenant since I’m not using keystone — how does it even know what tenant I’m in? | 15:01 |
| mugsie | if you use the no auth middleware, it injects a default tenant | 15:02 |
| mugsie | noauth-project | 15:02 |
| andrewbogott | ok | 15:02 |
| mugsie | you can inject your own via http headers if you want | 15:03 |
| mugsie | X-Auth-Project-ID:foo | 15:03 |
| andrewbogott | What does sink do? | 15:03 |
| mugsie | will set the tenant to foo | 15:03 |
| *** EricGonczer_ has quit IRC | 15:03 | |
| mugsie | eh... | 15:03 |
| *** EricGonczer_ has joined #openstack-dns | 15:03 | |
| *** pk_ has joined #openstack-dns | 15:04 | |
| andrewbogott | ‘eh…’ means you don’t know, or means it does the obvious right thing? :) | 15:04 |
| mugsie | i have no clue anymore ;) | 15:05 |
| mugsie | (i may be looking at sourcecode :/ ) | 15:05 |
| dvorak | vinod1: yeah, I thought it was really weird that they scheduled my talk opposite another designate talk | 15:05 |
| andrewbogott | I’m trying to backfill a bunch of pre-sink entries. Experiments suggest that sink can still delete them properly even if they’re installed without my specifying a tenant. So I will do my best not to worry about this :) | 15:06 |
| andrewbogott | I do have more than 500 of them though :( | 15:06 |
| mugsie | dvorak: they tend to have sections of days for certain topics | 15:06 |
| mugsie | andrewbogott: yeah, currently sink will | 15:06 |
| mugsie | but that is being depricated this cycle | 15:06 |
| mugsie | (it can be a pretty bad bug as it turns out) | 15:07 |
| dvorak | Just seems kind of silly to put two talks with almost exactly the same audience at the same time. | 15:07 |
| andrewbogott | Hm… | 15:07 |
| mugsie | i seems sink has no tenant id ? | 15:07 |
| mugsie | hummm | 15:07 |
| mugsie | I have never really done much with sink | 15:07 |
| andrewbogott | That sounds like I need to fix my backfill script to send the right tenant. | 15:07 |
| andrewbogott | Or else I can’t upgrade | 15:07 |
| mugsie | its the managed flag that needs to be set | 15:08 |
| mugsie | from the end of liberty sink will no longer be ablke top | 15:08 |
| andrewbogott | sure, I’m setting the managed flags and the associated instance. | 15:08 |
| mugsie | be able to interact with non managed=True records | 15:08 |
| andrewbogott | Just nothing about the tenant. | 15:08 |
| mugsie | then you shouod be fine | 15:08 |
| andrewbogott | great. | 15:08 |
| andrewbogott | I guess I need to dive into the source to figure out about this quota issue. Would you guess that that’s being enforced by the api? | 15:09 |
| mugsie | no, it should be done in central | 15:11 |
| andrewbogott | ok | 15:11 |
| mugsie | i think the tenant id for sink records may actually be none :S | 15:11 |
| * andrewbogott inserts a million debug lines | 15:11 | |
| mugsie | Kiall: any idea what the tenant id for sink records is? | 15:11 |
| dvorak | mugsie: I was looking at it this weekend, I think you can pass it using the rpc api | 15:12 |
| dvorak | I haven't written the sink for us yet, but my plan was hoping the notifications from neutron would have a tenant id in them that I could just pass through in the create request | 15:13 |
| mugsie | dvorak: oh, you can | 15:13 |
| mugsie | but in the default sample ones we ship it seems to be non | 15:13 |
| mugsie | none* | 15:13 |
| dvorak | nod | 15:13 |
| openstackgerrit | Merged openstack/designate: Refacor function for parsing host:port strings https://review.openstack.org/166322 | 15:14 |
| mugsie | at least in icehouse | 15:14 |
| openstackgerrit | Merged openstack/designate: Combine Backend and PoolBackend https://review.openstack.org/166609 | 15:15 |
| *** pk_ has quit IRC | 15:17 | |
| timsim | mugsie: Is it not the config value: managed_resource_tenant_id | 15:17 |
| mugsie | i think it should be | 15:19 |
| mugsie | but I dont see where it is getting set | 15:19 |
| mugsie | timsim: https://github.com/openstack/designate/blob/stable/icehouse/designate/notification_handler/base.py#L120 | 15:19 |
| timsim | Ah. /shrug | 15:22 |
| mugsie | timsim: :D | 15:23 |
| mugsie | I was just looking at it going "really??" | 15:23 |
| timsim | Does it fall back to that tenant or something? It looks like that would have tenant=None | 15:25 |
| *** jmcbride has joined #openstack-dns | 15:26 | |
| mugsie | yeah... | 15:30 |
| mugsie | it might be in storage i suppose | 15:30 |
| *** andrewbogott is now known as andrewbogott_afk | 15:30 | |
| mugsie | https://github.com/openstack/designate/search?utf8=%E2%9C%93&q=managed_resource_tenant_id | 15:30 |
| mugsie | doesnt look like it | 15:30 |
| *** rmoe has quit IRC | 15:31 | |
| *** andrewbogott_afk is now known as andrewbogott | 15:33 | |
| *** kodoku has quit IRC | 15:34 | |
| andrewbogott | mugsie: hm, I’m back where I was yesterday… when I grep for quota-recordset-records I find the place where the option is defined and the string recurs nowhere else in the source. | 15:40 |
| Kiall | I'm 99% sure that quota (quota-recordset-records) didn't exist in Icehouse? | 15:43 |
| Kiall | (on a call, didn't read scrollback, sorry if I'm way off :P) | 15:44 |
| *** rmoe has joined #openstack-dns | 15:50 | |
| andrewbogott | Kiall: My main issue is that I’m hitting a limit of 500 records (probably quota_domain_records) even though I have it set much higher in the config. Probably i’m misunderstanding how that global config setting relates to tenant-specific settings… | 15:53 |
| andrewbogott | But my ‘quotas’ table in the designate db is empty | 15:54 |
| Kiall | and .. deleted the entire message I was typing. heh | 15:54 |
| Kiall | sorry - still on a call, back and forth ;) | 15:56 |
| andrewbogott | np, I’m not stumped yet :) | 15:56 |
| Kiall | Anyway.. the config values are the defaults for when a tenant-specific quota isn't set.. So, if those are set higher than the practical limit your seeing, it should mean you have a tenant specific quota value in the DB | 15:56 |
| andrewbogott | Maybe I’m just setting the config in the wrong place. They should be in the ‘central’ section? | 15:58 |
| *** pk_ has joined #openstack-dns | 15:59 | |
| *** pk_ has quit IRC | 15:59 | |
| Kiall | andrewbogott: ah, there we go.. I believe they live in the [DEFAULT] section | 16:00 |
| Kiall | Though, that's probably the wrong place for them.. [DEFAULT] has a little bit of a random collection of "stuff" | 16:00 |
| andrewbogott | I’ve now copied them into every section, including one of my own invention… still no dice | 16:01 |
| timsim | andrewbogott: Do you have them as quota-domains or quota_domains? The underscore matters (I think :P) | 16:02 |
| * andrewbogott tries both | 16:03 | |
| timsim | They should be underscored, didn't make that clear. | 16:03 |
| andrewbogott | timsim: oh, even though they’re defined with - in the source? | 16:04 |
| andrewbogott | dammit, that was it | 16:04 |
| timsim | Yep. I think it's an oslo config thing. | 16:04 |
| andrewbogott | So… since those settings aren’t documented I had to pull them from the source, and in the source they are magically transformed in name… | 16:04 |
| * andrewbogott sighs | 16:04 | |
| andrewbogott | anyway, thank you! I’m off to the races :) | 16:05 |
| timsim | Have fun. | 16:05 |
| jschwarz | hi guys :) | 16:20 |
| jschwarz | So I've finally set up designate with the various agents and power DNS, and got my machine to respond to a 'nova1.example.com' DNS request | 16:23 |
| jschwarz | this is an all-in-one setup, so the same machine which can respond to DNS requests also has a few instances, tenant networks, etc | 16:24 |
| jschwarz | to enable my instances query the DNS server i had to connect the instances to the machine's LAN (since the pdns client is listening on a non-tenant network IP) | 16:25 |
| jschwarz | is there some way to connect the pdns to a tenant network directly (ie. connect the pdns server directly to the namespace?) | 16:25 |
| jschwarz | anyone? :< | 16:49 |
| elarson | jschwarz: sorry, what was the question? | 16:50 |
| jschwarz | is there some way to connect the pdns to a tenant network directly (ie. connect the pdns server directly to the namespace?) | 16:50 |
| elarson | jschwarz: so you configure your pdns (which I'm assuming is power dns) via designate, and then you want the specific tenent to use that configured pdns server? | 16:52 |
| jschwarz | elarson, exactly | 16:52 |
| elarson | jschwarz: it seems like that might be outside the bounds of what designate is meant to do insofar as it doesn't provide a means of configuring a pool of servers using a specific dns server, for example. | 16:53 |
| elarson | but I could be wrong! | 16:53 |
| jschwarz | elarson, uh.. so what does designate meant to do? | 16:54 |
| * jschwarz feels like that was a dumb question | 16:55 | |
| *** jordanP has quit IRC | 16:56 | |
| elarson | jschwarz: it provides an API to configure and manage DNS servers like bind9 and pdns. it sounds like what you want to do is configure your hosts to use a specific dns servers, which is designate doesn't do. | 16:56 |
| timsim | Wait, so you just want a single tenant to use that powerdns server, and others to use a different one? | 16:56 |
| jschwarz | timsim, yeah. It seems rather natural to be able to connect designate-controlled backends to tenant networks | 16:57 |
| * elarson hopes my description of what designate does is reasonably correct ;) | 16:57 | |
| jschwarz | timsim, looking at the nova-neutron-instances point of view, at least | 16:57 |
| elarson | jschwarz: if a tenant has a network already, what's to stop you from running designate within that network for configuring that network's dns? | 16:58 |
| * elarson is just thinking aloud | 16:58 | |
| jschwarz | elarson, I don't want to set-up 2 different openstack setups - one that hosts the tenants, networks, etc, and another that hosts designate for a single tenant | 16:59 |
| *** pk_ has joined #openstack-dns | 16:59 | |
| jschwarz | also, what if I have more than one tenant which could use a DNSaaS? | 16:59 |
| elarson | jschwarz: fair enough, just throwing it out there. | 17:00 |
| jschwarz | elarson, I think I completely missed the point of this project's effort ;-) | 17:00 |
| timsim | jschwarz: So Designate is multi-tenanted by default. But what it sounds like you want is for each tenant to have a different "pool" of dns server(s) is that a fair description? | 17:00 |
| * elarson steps out for some food | 17:00 | |
| jschwarz | timsim, sounds accurate enough | 17:01 |
| jschwarz | timsim, end-goal is having a tenant network be served my a pdns instance running by designate | 17:01 |
| *** untriaged-bot has joined #openstack-dns | 17:02 | |
| untriaged-bot | Untriaged bugs so far: | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1434189 | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1434479 | 17:02 |
| openstack | Launchpad bug 1434189 in Designate "Delete zone process can be interrupted" [Undecided,In progress] - Assigned to Tim Simmons (tim-simmons-t) | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432842 | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432850 | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1425668 | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1432839 | 17:02 |
| openstack | Launchpad bug 1434479 in Designate "designate stop working after insert ~10k records" [Critical,New] | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1433645 | 17:02 |
| openstack | Launchpad bug 1432842 in Designate "Property setters dont work." [Undecided,New] | 17:02 |
| openstack | Launchpad bug 1432850 in Designate "Add periodic refresh task" [High,New] - Assigned to Endre Karlson (endre-karlson) | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1430976 | 17:02 |
| openstack | Launchpad bug 1425668 in Designate "Poor error message when using same database for designate and the pool manager cache" [Undecided,New] | 17:02 |
| openstack | Launchpad bug 1432839 in Designate "Calling find_domain in any v1 operation is sub-optimal" [Medium,New] - Assigned to Endre Karlson (endre-karlson) | 17:02 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1425117 | 17:02 |
| openstack | Launchpad bug 1433645 in Designate "Floating IP Reverse DNS API does not respect async operations" [Undecided,New] | 17:02 |
| *** untriaged-bot has quit IRC | 17:02 | |
| openstack | Launchpad bug 1430976 in Designate "Pool Manager creates/deletes can go into an infinite loop" [Undecided,Fix committed] - Assigned to Tim Simmons (tim-simmons-t) | 17:02 |
| openstack | Launchpad bug 1425117 in Designate "Designate does not work with postgres" [High,New] | 17:02 |
| jschwarz | timsim, what do you mean by 'Designate is multi-tenanted by default'? | 17:03 |
| timsim | jschwarz: The default way to operating Designate is that you have it manage some set of DNS servers, and every tenant that creates DNS entries goes to those same servers, but Designate does some nice things to make sure that runs smoothly. | 17:04 |
| jschwarz | timsim, can that tenant's instances access those servers? | 17:05 |
| timsim | Sure, in most cases they probably don't run on a specific network, but for the openstack setup as a whole. | 17:06 |
| *** dan has quit IRC | 17:07 | |
| timsim | I'm heading out for lunch, I'll be back in a bit | 17:08 |
| jschwarz | timsim, Neutron allows one to create networks which are separated from the 'management' and 'external' networks, and can use tunneling/software routing to enable instance's connectivity to 8.8.8.8, for example | 17:08 |
| timsim | Right, but for resolving DNS for those instances, wouldn't you want those DNS servers on an external network so that clients not on that neutron net can resolve them? | 17:10 |
| jschwarz | but not necessarily - the networks can be completely hidden from the outside, and then they could use DNSaaS. or even they can have external connectivity but not to the management network (which is where my pdns is running, afaik) | 17:10 |
| timsim | You could still create dns records for those IPs, even if the DNS server itself has no access to them. | 17:11 |
| jschwarz | but then how could the instances use the DNS records if they can't access the DNS server? | 17:11 |
| timsim | Well the instances should be able to if that DNS server is running on some more public network that all of the instances have access to. | 17:12 |
| *** dan has joined #openstack-dns | 17:12 | |
| timsim | Anyway, I'll be back in a bit, need sustenance. | 17:13 |
| jschwarz | enjoy | 17:13 |
| ekarlso | hmmm timsim | 17:13 |
| *** vinod1 has quit IRC | 17:14 | |
| ekarlso | any clues on why when mass creating recordsets sequentially pool manage queues stagger up ? | 17:14 |
| jschwarz | That's the idea - you should be able to 'plug' the DNS server to some network/namespace of your choosing, and then the instances will have connectivity to it without requiring the private network to be connected to some 'management' network (which is for OpenStack management, not tenant data) | 17:14 |
| jschwarz | Currently (as far as I can tell), you can only run the DNS servers in the management network, which isn't always connected to the tenants at all | 17:15 |
| jschwarz | (why should tenants have access to OpenStack's API servers?) | 17:15 |
| mugsie | jschwarz: we are location agnositic for the DNS servers | 17:16 |
| mugsie | you use case is a valid one, and it is on our roadmap | 17:16 |
| mugsie | you're | 17:16 |
| mugsie | * | 17:16 |
| jschwarz | mugsie, excellent :) | 17:16 |
| mugsie | but we need to get tighter integration with neutron first | 17:16 |
| jschwarz | mugsie, where can I see that roadmap and how can i contribute to it? :) | 17:16 |
| mugsie | there is blueprints, but not sure they are there - we need to get them up to date | 17:17 |
| mugsie | most they are talked about at midcycles and summits | 17:17 |
| jschwarz | I see | 17:18 |
| jschwarz | do you have a time estimation? | 17:18 |
| jschwarz | (a cycle? 2?) | 17:18 |
| mugsie | 1-2 cycles | 17:18 |
| mugsie | minimum | 17:18 |
| mugsie | but - there should be featres that you could roll your own before | 17:18 |
| jschwarz | mugsie, sounds like words of wisdom and pain :) | 17:19 |
| mugsie | just not as nicely as we could do it otherwise | 17:19 |
| mugsie | yeah... | 17:19 |
| jschwarz | well, I'm interested in helping out a bit | 17:20 |
| mugsie | cool. | 17:20 |
| mugsie | when we get out of rc mode, I will start back on documenting the future :) | 17:20 |
| mugsie | also - speaking from experience - please allow VMs to access OpenStack APIs ;) | 17:21 |
| jschwarz | mugsie, alas, not all deployment can allow that | 17:21 |
| mugsie | it is a royal pain when you can't get access from a VM, depending on what you re doing | 17:21 |
| mugsie | are* | 17:21 |
| mugsie | yeah... makes life .... interesting .... | 17:21 |
| jschwarz | mugsie, my all-in-one doesn't really care, but when you're talking about a dozen of machines... | 17:22 |
| jschwarz | interesting is another word for that | 17:22 |
| jschwarz | mugsie, you were talking about a tighter integration with neutron? | 17:22 |
| jschwarz | in what regard? | 17:22 |
| mugsie | so when yoiu create a new port on the network, it calls out to us and creates records for the IPs associated with it | 17:24 |
| mugsie | and the ablity to create ports on tenant networks without designate needing admin creds | 17:24 |
| mugsie | but that last one should fixed afaik | 17:24 |
| jschwarz | the first one isn't implemented yet? | 17:24 |
| mugsie | nope | 17:25 |
| mugsie | we listen to events from the notifcatiojn handler | 17:25 |
| mugsie | but we need to move to an active call out from neutron | 17:25 |
| *** richm has quit IRC | 17:25 | |
| jschwarz | mugsie, sounds nice. anything else? | 17:27 |
| mugsie | forcing the dnsmasq to forward queries for certain domains to our resolver | 17:27 |
| jschwarz | sounds reasonable, though for that I think that the resolved should be able to be plugged to the same network | 17:28 |
| jschwarz | no? | 17:28 |
| *** richm has joined #openstack-dns | 17:30 | |
| jschwarz | right, that's it for today. Thanks for the help guys :) | 17:39 |
| mugsie | jschwarz: yeah - its all a thought excersise right now though :) | 17:40 |
| mugsie | see ya later o/ | 17:40 |
| jschwarz | :) | 17:40 |
| ekarlso | funny thing | 17:41 |
| ekarlso | if you have 1000 record creations atm | 17:41 |
| ekarlso | you'll bork up designate badly :P | 17:41 |
| *** jschwarz has quit IRC | 17:45 | |
| *** pk_ has quit IRC | 17:58 | |
| *** GonZo2K has joined #openstack-dns | 18:02 | |
| *** pk_ has joined #openstack-dns | 18:07 | |
| *** dan has quit IRC | 18:08 | |
| *** vinod1 has joined #openstack-dns | 18:11 | |
| *** dank_ has joined #openstack-dns | 18:14 | |
| *** pk_ has quit IRC | 18:18 | |
| *** pk_ has joined #openstack-dns | 18:18 | |
| timsim | ekarlso: The Pool Manager stacks up? | 18:19 |
| *** pk_ has quit IRC | 18:19 | |
| *** pk_ has joined #openstack-dns | 18:20 | |
| openstackgerrit | Merged openstack/designate: Archive non-pool backends https://review.openstack.org/166610 | 18:20 |
| *** pk_ has quit IRC | 18:21 | |
| ekarlso | timsim: central actually I think :p | 18:22 |
| ekarlso | but it's a sqla pool thing atm it seems | 18:22 |
| *** pk_ has joined #openstack-dns | 18:22 | |
| timsim | That was in my plan to test in the next few weeks. | 18:23 |
| *** pk__ has joined #openstack-dns | 18:23 | |
| *** pk_ has quit IRC | 18:23 | |
| Kiall | timsim: https://github.com/openstack/designate/blob/master/designate/central/service.py#L2099 is selecting and updating WAAYYY too much data on every zone change | 18:24 |
| Kiall | (That's the stuff endre spotted..) | 18:24 |
| ekarlso | Kiall: not way too much... way too much to make the whole central blow up... | 18:26 |
| timsim | Ah yeah...updating the status of every single record when there's thousands of them. Maybe not the best idea. | 18:26 |
| ekarlso | central just gets clogged... | 18:26 |
| ekarlso | if that gets into kilo having large zones will be useless :§ | 18:26 |
| timsim | That's a big select and then a query per record. | 18:26 |
| ekarlso | yep :D | 18:27 |
| timsim | woops | 18:27 |
| ekarlso | timsim: basiaclly my devstack central just keeps doing update_record_status or so infinelety :P | 18:27 |
| ekarlso | pool manager does 1 call then ^ happens | 18:27 |
| ekarlso | pretty (d0h) | 18:27 |
| timsim | Maybe only update the SOA record when you have a pending domain? Or the individual records that are actually changing? | 18:27 |
| *** pk__ has quit IRC | 18:28 | |
| ekarlso | timsim: i dunno, I just know that it blows the entire system up if you have a large zone :p | 18:28 |
| ekarlso | but yeah, ^ sounds like agood idea | 18:28 |
| *** pk_ has joined #openstack-dns | 18:35 | |
| *** pk_ has quit IRC | 18:39 | |
| *** pk_ has joined #openstack-dns | 18:42 | |
| *** ericpeterson has joined #openstack-dns | 18:48 | |
| ekarlso | hey EricGonczer_ :) | 18:48 |
| ekarlso | ehm, ericpeterson | 18:48 |
| ericpeterson | hey. questions with https://github.com/openstack/designate/tree/master/contrib/designate-dashboard | 18:50 |
| ericpeterson | should I use gerrit and launchpad bugs etc if I have changes to propose / share? | 18:50 |
| ericpeterson | so far, the changes I have are around packaging / deployment stuff, probably around the setup.cfg file within there. thanks!! :D | 18:51 |
| mugsie | ericpeterson: yup - that would be great | 18:57 |
| mugsie | we use launchpad / gerrit for that folder as well | 18:57 |
| mugsie | it does need a bit of attention :) | 18:58 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Utility script to help create n+ records at a time https://review.openstack.org/166967 | 18:58 |
| ekarlso | boris-42: did you ever get to implement osprofiler to designate ? | 19:00 |
| boris-42 | ekarlso: there are patches on review | 19:02 |
| boris-42 | ekarlso: but there are some issues | 19:02 |
| *** ericpeterson has quit IRC | 19:03 | |
| *** raginbajin has quit IRC | 19:04 | |
| *** raginbajin has joined #openstack-dns | 19:04 | |
| *** mikal has quit IRC | 19:04 | |
| *** mikal has joined #openstack-dns | 19:04 | |
| ekarlso | boris-42: crap :p | 19:07 |
| ekarlso | timsim: so yeah, about 200 records now in the db (upping it gradually to see where it dies) and each update call from pm to central is taking about 5-6 secs | 19:08 |
| timsim | Yeah, gotta be all those individual record updates. | 19:08 |
| ekarlso | :P | 19:08 |
| boris-42 | ekarlso: so run it in gatees! | 19:09 |
| boris-42 | =) | 19:09 |
| ekarlso | boris-42: first I wanna find out when it breaks :/ | 19:10 |
| ekarlso | boris-42: it's two issues it seems https://bugs.launchpad.net/designate/+bug/1434479 is one with pooling issues and the other is the updates we figure | 19:11 |
| openstack | Launchpad bug 1434479 in Designate "designate stop working after insert ~10k records" [Critical,New] | 19:11 |
| *** ericpeterson has joined #openstack-dns | 19:14 | |
| boris-42 | ekarlso: so you can just push patch on reviw | 19:16 |
| boris-42 | ekarlso: and continue working on your local env | 19:16 |
| boris-42 | ekarlso: why it's hard to push the patch ?) | 19:16 |
| *** pk_ has quit IRC | 19:19 | |
| *** pk_ has joined #openstack-dns | 19:24 | |
| *** ericpeterson has quit IRC | 19:24 | |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add API_v2 Adapter Objects https://review.openstack.org/164327 | 19:30 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Added FloatingIP object to DesignateObjects https://review.openstack.org/165453 | 19:30 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Validation Error Middleware does translation of keys to new format https://review.openstack.org/166982 | 19:30 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Replace API_v2 views with Adapters https://review.openstack.org/166983 | 19:30 |
| *** pk_ has quit IRC | 19:34 | |
| mugsie | vinod1: timsim Kiall ^^ are important :) | 19:35 |
| *** pk_ has joined #openstack-dns | 19:35 | |
| *** ericpeterson has joined #openstack-dns | 19:35 | |
| *** andrewbogott is now known as andrewbogott_afk | 19:38 | |
| vinod1 | will look at them mugsie | 19:38 |
| *** andrewbogott_afk is now known as andrewbogott | 19:39 | |
| mugsie | vinod1: ty | 19:39 |
| *** pk_ has quit IRC | 19:40 | |
| *** jmcbride has quit IRC | 19:55 | |
| vinod1 | mugsie: I had a question on https://review.openstack.org/#/c/165453/3/designate/objects/floating_ip.py | 20:11 |
| mugsie | not allowed | 20:12 |
| mugsie | :) | 20:12 |
| mugsie | sure - ask away | 20:12 |
| *** jmcbride has joined #openstack-dns | 20:12 | |
| vinod1 | ptrdname previously had the format hostname - https://github.com/openstack/designate/blob/master/designate/resources/schemas/v2/floatingip.json | 20:12 |
| *** jmcbride has quit IRC | 20:12 | |
| vinod1 | Now it is domainname | 20:12 |
| *** jmcbride has joined #openstack-dns | 20:12 | |
| vinod1 | just wanted to be sure that this change does not cause any problems | 20:12 |
| mugsie | nope - it shouldn't - ptrs need to full domainnames | 20:13 |
| vinod1 | also the ttl minimum is now 1 instead of 0 | 20:14 |
| mugsie | yes, we should be enforcing that for user input | 20:15 |
| vinod1 | cool - thanks for the clarifications - +A | 20:16 |
| mugsie | :D | 20:17 |
| *** jmcbride1 has joined #openstack-dns | 20:22 | |
| *** crc32 has joined #openstack-dns | 20:22 | |
| *** jmcbride has quit IRC | 20:22 | |
| *** crc32 has quit IRC | 20:23 | |
| timsim | mugsie: I get a schema error on create zone: http://paste.openstack.org/show/195555/ | 20:29 |
| timsim | (Testing the whole set of changes) | 20:29 |
| *** pk__ has joined #openstack-dns | 20:35 | |
| *** crc32 has joined #openstack-dns | 20:36 | |
| mugsie | timsim: damn it | 20:38 |
| mugsie | we dont really need ttls do we? | 20:38 |
| mugsie | :D | 20:38 |
| *** pk__ has quit IRC | 20:40 | |
| timsim | Nah. Just set it all to 1 sec, DNS changes should be instant :P | 20:40 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add API_v2 Adapter Objects https://review.openstack.org/164327 | 20:46 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Replace API_v2 views with Adapters https://review.openstack.org/166983 | 20:46 |
| *** mwagner_lap has quit IRC | 20:47 | |
| timsim | mugsie: Should "masters" on that one be read-only: False as well? | 20:49 |
| *** vipuls is now known as vipul | 20:49 | |
| *** pk__ has joined #openstack-dns | 20:50 | |
| mugsie | timsim: not at the moment - due to the weird way they work, I it done in the overloaded method below :( | 20:51 |
| mugsie | it is * | 20:51 |
| timsim | Oh yeah. Makes sense | 20:51 |
| mugsie | for the limits one.... I am of 2 minds | 20:52 |
| mugsie | not sure if the non wrapper one is right, or a list style wrapper | 20:53 |
| timsim | Ah. | 20:53 |
| mugsie | what do you think? | 20:53 |
| timsim | When you put it that way, I think I prefer the way it is now. | 20:54 |
| mugsie | :) | 20:54 |
| timsim | Because they are separate limits, but you can't have one without the others. It's an "absolute limit" object with attributes. | 20:55 |
| timsim | which happen to be separate limits. | 20:55 |
| timsim | If we wanted to add other limits that weren't absolute limits, that might be weird though. | 20:56 |
| *** vinod1 has quit IRC | 20:56 | |
| openstackgerrit | Graham Hayes proposed openstack/designate: Fixed miss-spellings in central https://review.openstack.org/167007 | 20:56 |
| mugsie | timsim: yeah. | 20:57 |
| mugsie | i think we can leave it as is, and if we need extra limits, we can add a sub controller | 20:57 |
| mugsie | also ^ ;) | 20:58 |
| timsim | Fair enough | 20:58 |
| timsim | Ironic misspelling of misspellings :) | 20:58 |
| *** vinod1 has joined #openstack-dns | 20:59 | |
| openstackgerrit | Graham Hayes proposed openstack/designate: Replace API_v2 views with Adapters https://review.openstack.org/166983 | 21:09 |
| openstackgerrit | Merged openstack/designate: Imported Translations from Transifex https://review.openstack.org/166506 | 21:10 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Fixed misspellings in central https://review.openstack.org/167007 | 21:10 |
| timsim | lol you didn't have to fix it ;) | 21:11 |
| mugsie | it was right there :) | 21:13 |
| *** EricGonczer_ has quit IRC | 21:38 | |
| *** jmcbride1 has quit IRC | 21:39 | |
| *** jmcbride has joined #openstack-dns | 21:40 | |
| *** jmcbride has quit IRC | 22:08 | |
| openstackgerrit | Merged openstack/designate: Fixed misspellings in central https://review.openstack.org/167007 | 22:13 |
| mugsie | ericpeterson: i like the bug description :) | 22:15 |
| ericpeterson | yeah, sorry if it is not very descriptive | 22:15 |
| ericpeterson | I have a change almost ready, working on docs for it | 22:16 |
| ericpeterson | :) | 22:16 |
| *** MentalRay has joined #openstack-dns | 22:16 | |
| mugsie | nope - it sums up the current situation perfectly ;) | 22:16 |
| openstackgerrit | Eric Peterson proposed openstack/designate: improving designatedashboard packaging https://review.openstack.org/167024 | 22:23 |
| openstackgerrit | Eric Peterson proposed openstack/designate: improving designatedashboard packaging https://review.openstack.org/167024 | 22:24 |
| ericpeterson | missed a trailing space :\ | 22:24 |
| openstackgerrit | Merged openstack/designate: Added FloatingIP object to DesignateObjects https://review.openstack.org/165453 | 22:31 |
| openstackgerrit | Merged openstack/designate: Validation Error Middleware does translation of keys to new format https://review.openstack.org/166982 | 22:32 |
| *** MentalRay has quit IRC | 22:36 | |
| *** jmcbride has joined #openstack-dns | 22:51 | |
| *** vinod1 has quit IRC | 22:51 | |
| *** paul_glass has quit IRC | 22:53 | |
| *** crc32 has quit IRC | 22:53 | |
| *** andrewbogott is now known as andrewbogott_afk | 23:06 | |
| *** EricGonczer_ has joined #openstack-dns | 23:22 | |
| *** ericpeterson has quit IRC | 23:26 | |
| *** pk__ has quit IRC | 23:27 | |
| *** chlong has joined #openstack-dns | 23:31 | |
| *** pk has joined #openstack-dns | 23:33 | |
| *** mwagner_lap has joined #openstack-dns | 23:34 | |
| *** EricGonc_ has joined #openstack-dns | 23:45 | |
| *** EricGonczer_ has quit IRC | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!