| *** openstack has joined #openstack-dev | 16:52 | |
| *** dolphm has joined #openstack-dev | 16:53 | |
| *** Remco__ has quit IRC | 16:54 | |
| antonym | vishy: it's back up, i ping'd monty about it | 16:55 |
|---|---|---|
| vishy | antonym: cool thx | 16:55 |
| *** ohnoimdead has joined #openstack-dev | 16:56 | |
| *** zns has quit IRC | 16:56 | |
| *** martine has joined #openstack-dev | 16:59 | |
| mtaylor | pvo: thanks bro! | 17:00 |
| pvo | mtaylor: no worries. | 17:01 |
| mtaylor | Daviey: r'\b([Bb]ug|[Ll][Pp])\s*[#:]?\s*(\d+)' | 17:01 |
| openstackgerrit | gholt proposed a change to openstack/swift: Fix bug regarding negative Content-Length headers. https://review.openstack.org/3581 | 17:02 |
| mtaylor | xtoddx: unabandoned | 17:03 |
| LinuxJedi | xtoddx: you fee is one pint of beer to mtaylor ;) | 17:03 |
| LinuxJedi | s/you/your | 17:03 |
| Daviey | thanks mtaylor | 17:04 |
| xtoddx | can do | 17:05 |
| xtoddx | mtaylor: thanks | 17:05 |
| *** spiffxp has joined #openstack-dev | 17:06 | |
| *** dwalleck has quit IRC | 17:06 | |
| *** maploin has quit IRC | 17:10 | |
| *** martine has quit IRC | 17:11 | |
| *** zns has joined #openstack-dev | 17:16 | |
| *** hattwick_ has joined #openstack-dev | 17:17 | |
| *** Mandell_ has joined #openstack-dev | 17:17 | |
| *** rods has quit IRC | 17:17 | |
| *** aclark_ has joined #openstack-dev | 17:18 | |
| *** vish1 has joined #openstack-dev | 17:19 | |
| *** cmagina_ has joined #openstack-dev | 17:19 | |
| *** mtaylor_ has joined #openstack-dev | 17:21 | |
| *** cp16net_ has joined #openstack-dev | 17:24 | |
| *** andrewbogott has joined #openstack-dev | 17:24 | |
| *** openstackgerrit has quit IRC | 17:24 | |
| *** markvoelker has quit IRC | 17:24 | |
| *** hattwick has quit IRC | 17:24 | |
| *** hattwick_ is now known as hattwick | 17:24 | |
| *** Mandell has quit IRC | 17:24 | |
| *** vishy has quit IRC | 17:24 | |
| *** AlanClark has quit IRC | 17:24 | |
| *** mtaylor has quit IRC | 17:24 | |
| *** cmagina has quit IRC | 17:24 | |
| *** ipl31_ has joined #openstack-dev | 17:24 | |
| *** ipl31 has quit IRC | 17:24 | |
| *** cp16net has quit IRC | 17:24 | |
| *** russellb has quit IRC | 17:24 | |
| *** cp16net_ is now known as cp16net | 17:24 | |
| *** russellb has joined #openstack-dev | 17:25 | |
| *** dtroyer has joined #openstack-dev | 17:26 | |
| *** mattstep has quit IRC | 17:26 | |
| *** novas0x2a|laptop has quit IRC | 17:29 | |
| *** openstackgerrit has joined #openstack-dev | 17:32 | |
| *** adjohn has joined #openstack-dev | 17:33 | |
| *** jog0 has joined #openstack-dev | 17:33 | |
| *** vincentricci has joined #openstack-dev | 17:33 | |
| *** mattstep has joined #openstack-dev | 17:33 | |
| *** adjohn has quit IRC | 17:33 | |
| *** adjohn has joined #openstack-dev | 17:33 | |
| openstackjenkins | Project nova-docs build #2074: SUCCESS in 3 min 17 sec: https://jenkins.openstack.org/job/nova-docs/2074/ | 17:33 |
| openstackjenkins | Hengqing Hu: Return instancesSet in TerminateInstances ec2 api | 17:33 |
| *** markvoelker has joined #openstack-dev | 17:34 | |
| *** aclark_ has quit IRC | 17:37 | |
| *** derekh has quit IRC | 17:39 | |
| *** jog0_ has joined #openstack-dev | 17:40 | |
| *** aclark_ has joined #openstack-dev | 17:41 | |
| openstackgerrit | Josh Kearney proposed a change to openstack/python-novaclient: Allow the service type to be specified via CLI or env var. https://review.openstack.org/3560 | 17:42 |
| openstackgerrit | Verification of a change to openstack/nova failed: bug 923798: On XenServer the DomU firewall driver fails with NotImplementedError https://review.openstack.org/3580 | 17:42 |
| uvirtbot | Launchpad bug 923798 in nova "On XenServer the DomU firewall driver fails with NotImplementedError" [Undecided,In progress] https://launchpad.net/bugs/923798 | 17:42 |
| *** jog0 has quit IRC | 17:42 | |
| *** jog0_ is now known as jog0 | 17:42 | |
| *** dwalleck has joined #openstack-dev | 17:43 | |
| *** bepernoot has joined #openstack-dev | 17:45 | |
| bodepd | does it make sense to supply a region to requests to the nova ec2 api, it seems like it means something a little different than it does in aws? | 17:47 |
| *** dolphm has quit IRC | 17:48 | |
| *** dolphm has joined #openstack-dev | 17:49 | |
| *** hashar has left #openstack-dev | 17:51 | |
| *** dwalleck has quit IRC | 17:54 | |
| *** reed has joined #openstack-dev | 17:55 | |
| *** jog0 has quit IRC | 17:55 | |
| *** rods has joined #openstack-dev | 17:55 | |
| *** jog0 has joined #openstack-dev | 17:55 | |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Be more explicit about emptying connection pool. https://review.openstack.org/3583 | 17:56 |
| *** gyee has joined #openstack-dev | 17:57 | |
| *** bepernoot has quit IRC | 17:57 | |
| *** jdurgin has joined #openstack-dev | 17:57 | |
| *** jog0 has quit IRC | 17:58 | |
| *** jog0 has joined #openstack-dev | 17:59 | |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Clear out RPC connection pool before exit. https://review.openstack.org/3584 | 18:00 |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Empty connection pool after test_kombu. https://review.openstack.org/3585 | 18:00 |
| *** mtaylor_ is now known as mtaylor | 18:01 | |
| *** mtaylor has quit IRC | 18:01 | |
| *** mtaylor has joined #openstack-dev | 18:01 | |
| *** ChanServ sets mode: +v mtaylor | 18:01 | |
| ayoung | termie, can you join the keystone meeting so we can discus ksl? | 18:03 |
| *** joesavak has joined #openstack-dev | 18:04 | |
| *** stuntmac_ has joined #openstack-dev | 18:04 | |
| *** stuntmachine has quit IRC | 18:06 | |
| *** rods has quit IRC | 18:18 | |
| *** dolphm has quit IRC | 18:19 | |
| *** dolphm has joined #openstack-dev | 18:19 | |
| *** heckj has joined #openstack-dev | 18:21 | |
| *** dolphm has quit IRC | 18:23 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:25 | |
| openstackgerrit | Dean Troyer proposed a change to openstack-dev/devstack: Add support for OS_* environment vars https://review.openstack.org/3493 | 18:27 |
| openstackgerrit | eglynn proposed a change to openstack/glance: Add --capture-output option to glance-control. https://review.openstack.org/3516 | 18:27 |
| *** darraghb has quit IRC | 18:28 | |
| *** dolphm has joined #openstack-dev | 18:29 | |
| openstackgerrit | Dean Troyer proposed a change to openstack-dev/devstack: Extract common functions into a separate file https://review.openstack.org/3490 | 18:29 |
| *** Ryan_Lane has joined #openstack-dev | 18:31 | |
| vish1 | comstud: ping | 18:31 |
| openstackjenkins | Project nova-docs build #2075: SUCCESS in 2 min 49 sec: https://jenkins.openstack.org/job/nova-docs/2075/ | 18:33 |
| openstackjenkins | * armando.migliaccio: bug 923798: On XenServer the DomU firewall driver fails with NotImplementedError | 18:33 |
| openstackjenkins | * jason: fixes melange ipam lib | 18:33 |
| uvirtbot | Launchpad bug 923798 in nova "On XenServer the DomU firewall driver fails with NotImplementedError" [Undecided,Fix committed] https://launchpad.net/bugs/923798 | 18:33 |
| openstackgerrit | eglynn proposed a change to openstack/glance: Add --capture-output option to glance-control. https://review.openstack.org/3516 | 18:36 |
| *** spiffxp has quit IRC | 18:37 | |
| *** spiffxp has joined #openstack-dev | 18:37 | |
| *** vincentricci_ has joined #openstack-dev | 18:40 | |
| *** vincentricci has quit IRC | 18:40 | |
| *** vincentricci_ is now known as vincentricci | 18:40 | |
| *** bepernoot has joined #openstack-dev | 18:42 | |
| comstud | vish1: pong | 18:48 |
| ogelbukh | hello, can someone point out what is the status of atlas-lb within the openstack? | 18:51 |
| *** vish1 is now known as vishy | 18:54 | |
| *** rbasak has quit IRC | 18:58 | |
| vishy | comstud: https://github.com/vishvananda/nova/compare/master...initiator#L9R348 | 18:59 |
| *** novas0x2a|laptop has quit IRC | 18:59 | |
| vishy | comstud: unfamiliar with xenapi host stats, but I think that is the right way to get a config setting? | 18:59 |
| *** novas0x2a|laptop has joined #openstack-dev | 18:59 | |
| *** anotherjesse has joined #openstack-dev | 19:00 | |
| heckj | young - you here? | 19:00 |
| heckj | ayoung^^ | 19:00 |
| ayoung | heckj, I'm here | 19:00 |
| heckj | ayoung: the KSL branch is more consolidated for those entities, not attempting to separate them for various endpoints. You'd create a different backend that used the pieces that you want as you want them there - so a single place coordinates the efforts between sql, ldap, etc. | 19:01 |
| zns | anotherjesse: how is this handled in ksl? ayoung said: "anotherjesse, do we keep the backend_entities keyword, and have the dirver honor that?" | 19:01 |
| *** davlap has joined #openstack-dev | 19:02 | |
| zns | heckj: so you use drivers for groups of entities (i.e. I have driver for tenants, roles, and users - but I can't put them on different drivers). Correct? | 19:02 |
| heckj | ayoung: It just requires you implement the API that's defined as an abstract base object (https://github.com/openstack/keystone/blob/redux/keystone/identity/core.py) - Driver | 19:02 |
| *** dubsquared1 has quit IRC | 19:03 | |
| anotherjesse | heckj / ayoung - there are (at least) two ways to do backing to multiple backends | 19:03 |
| heckj | zns: its more organized around the API that you interact with rather than the objects being stored. | 19:03 |
| anotherjesse | one idea is to specify each driver independently | 19:03 |
| anotherjesse | the other is that you build an identity driver that can put together the backends in a way that makes sense - but is wholesale configured | 19:04 |
| *** flaviamissi has quit IRC | 19:04 | |
| *** bepernoot has quit IRC | 19:05 | |
| heckj | anotherjesse: do we have anything of the second yet defined in KSL? | 19:05 |
| openstackjenkins | Project nova-docs build #2076: SUCCESS in 5 min 1 sec: https://jenkins.openstack.org/job/nova-docs/2076/ | 19:05 |
| openstackjenkins | rbryant: Be more explicit about emptying connection pool. | 19:05 |
| *** _adjohn has joined #openstack-dev | 19:06 | |
| *** flaviamissi has joined #openstack-dev | 19:07 | |
| anotherjesse | heckj: for composing? | 19:07 |
| heckj | anotherjesse: yeah | 19:07 |
| heckj | anotherjesse: an example of configuration based composing anyway | 19:07 |
| ayoung | anotherjesse, heckj OK, I need to start somehwhere. If I just start using the sql code as a prototype of how to write the LDAP one, I assume we can do the composing on top of that, maybe as a second step? | 19:07 |
| anotherjesse | ayoung: the kvs is probably simplier | 19:07 |
| ayoung | anotherjesse, OK | 19:08 |
| anotherjesse | but it is that you need to implement: | 19:08 |
| anotherjesse | def get_tenant(self, tenant_id): | 19:08 |
| anotherjesse | def get_tenant_by_name(self, tenant_name): | 19:08 |
| anotherjesse | ... | 19:08 |
| *** jog0 has quit IRC | 19:08 | |
| openstackgerrit | A change to openstack/nova has been rejected: Changes for supporting fast cloning on Xenserver. (Implements https://blueprints.launchpad.net/nova/+spec/fast-cloning-for-xenserver) 1. use_cow_images flag is reused for xenserver to check if copy on write images should be used. 2. image-id is used to https://review.openstack.org/3343 | 19:08 |
| ayoung | anotherjesse, and I've been told that there is code in Nova that we are looking at? | 19:08 |
| anotherjesse | ayoung: https://github.com/openstack/keystone/blob/redux/keystone/identity/core.py is actually the base class | 19:08 |
| *** jog0 has joined #openstack-dev | 19:09 | |
| anotherjesse | ayoung: https://github.com/openstack/nova/blob/master/nova/auth/ldapdriver.py | 19:09 |
| heckj | ayoung: the class "Driver" in there | 19:09 |
| comstud | vishy: looking | 19:09 |
| openstackgerrit | Mikyung Kang proposed a change to openstack/nova: Implements blueprint heterogeneous-tilera-architecture-support https://review.openstack.org/1402 | 19:09 |
| *** adjohn has quit IRC | 19:09 | |
| *** _adjohn is now known as adjohn | 19:09 | |
| ayoung | anotherjesse, heckj do we have an LDAP scheme we expect the dirsrv to have in it? I followed the steps http://mirantis.blogspot.com/2011/08/ldap-identity-store-for-openstack.html but had to guess at what was expected for the LDIF to define user, groups and so on | 19:10 |
| anotherjesse | ayoung: https://github.com/openstack/nova/blob/master/nova/auth/nova_openldap.schema | 19:11 |
| heckj | ayoung: that's been different for nearly every organization I've run into so far - configuring a LDAP search string in a config file has been my approach so far, but I don't have any public code to show you on that | 19:11 |
| vishy | zns: sudo security question | 19:11 |
| vishy | zns: oops wrong person | 19:11 |
| ayoung | anotherjesse, thanks | 19:12 |
| vishy | smoser: ping | 19:12 |
| smoser | hey | 19:12 |
| vishy | smoser: so if you had a choice | 19:12 |
| vishy | needing to escalate to root | 19:13 |
| *** jperkin has joined #openstack-dev | 19:13 | |
| vishy | cat or grep? | 19:13 |
| vishy | smoser: the deal is i need to read a conf file that is owned by root | 19:13 |
| ayoung | anotherjesse, heckj OK, I'll start with those. I'll bother you if I get stuck. | 19:13 |
| * heckj nods | 19:13 | |
| comstud | vishy: won't quite work... only certain information is returned from get_host_stats() it appears | 19:14 |
| vishy | comstud: darn, it looked like the config values were copied | 19:14 |
| comstud | vishy: See 'def update_status' in HostState class in xenapi_conn | 19:14 |
| *** martine has joined #openstack-dev | 19:15 | |
| comstud | vishy: oh wait | 19:15 |
| comstud | vishy: I missed the top part of that method | 19:15 |
| smoser | vishy, i dont know. cat seems less likely to be vulnerable to anything, and both that i see aren't going to open anything other than stdout for write. | 19:15 |
| *** vincentricci has quit IRC | 19:16 | |
| vishy | comstud: wait it looks like it doesn't copy the whole sr rec | 19:16 |
| comstud | vishy: yeah, not the whole SR record... but I see it copies the 'host_data' information | 19:17 |
| comstud | vishy: At the top part of update_status() | 19:17 |
| comstud | vishy: 660 data = json.loads(task_result) | 19:17 |
| comstud | vishy: I think that does give you what you need | 19:17 |
| comstud | vishy: But.. I don't know enough about Xen to say that 'host_other-config' is always going to be there or anything | 19:18 |
| vishy | comstud: ok right so i did read it correctly the first time | 19:18 |
| comstud | vishy: appears so :) | 19:18 |
| vishy | comstud: according to xen docs it should be set | 19:18 |
| vishy | but I will ask ewan | 19:18 |
| comstud | i'll look at my xen install | 19:18 |
| comstud | and see what comes back from that | 19:18 |
| comstud | if i can figure out the corresponding 'xe' command | 19:19 |
| comstud | oh | 19:20 |
| comstud | calls plugin | 19:20 |
| comstud | ok | 19:20 |
| *** jakedahn has joined #openstack-dev | 19:21 | |
| comstud | vishy: In my output, it shows the key as 'other-config' | 19:21 |
| vishy | comstud: 337 oc = dct.get("other-config", "") | 19:21 |
| vishy | 338 out["host_other-config"] = ocd = {} | 19:21 |
| vishy | in the plugin | 19:21 |
| vishy | are you getting it some other way? | 19:22 |
| vishy | or maybe it splits out host on the other end? | 19:22 |
| *** bepernoot has joined #openstack-dev | 19:22 | |
| comstud | vishy: ah, the plugin is translating the keys | 19:22 |
| comstud | vishy: i was doing a manual 'xe host-param-list' | 19:22 |
| *** nati2 has joined #openstack-dev | 19:22 | |
| vishy | comstud: cool, so that should work supposedly? | 19:23 |
| *** rbasak has joined #openstack-dev | 19:23 | |
| comstud | vishy: So I think you're set | 19:23 |
| comstud | vishy: Appears so | 19:23 |
| openstackgerrit | Verification of a change to openstack-dev/devstack failed: Add support for OS_* environment vars https://review.openstack.org/3493 | 19:23 |
| vishy | comstud: ok I'm going to prop this then | 19:24 |
| *** dolphm has quit IRC | 19:24 | |
| *** dwalleck has joined #openstack-dev | 19:24 | |
| comstud | vishy: cools. i'm dropping off shortly to head to airport. should have a reply about that FFE before then | 19:31 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Add initiator to initialize_connection https://review.openstack.org/3586 | 19:31 |
| vishy | comstud: great | 19:31 |
| *** pixelbeat has quit IRC | 19:32 | |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Add initiator to initialize_connection https://review.openstack.org/3586 | 19:37 |
| dtroyer | mtaylor: ping? | 19:41 |
| mtaylor | dtroyer: pong | 19:41 |
| dtroyer | mtaylor: got an interesting devstack test problem. | 19:41 |
| *** nati2 has quit IRC | 19:42 | |
| dtroyer | mtaylor: https://review.openstack.org/3493 adds an exercise (client-env.sh) that doesn't pass yet because code needs to be fixed | 19:42 |
| *** nati2 has joined #openstack-dev | 19:42 | |
| dtroyer | mtaylor: can you exempt it from running under jenkins for now? | 19:42 |
| *** jmolinaso has joined #openstack-dev | 19:42 | |
| mtaylor | you mean just the client-env.sh ? | 19:43 |
| dtroyer | yes | 19:43 |
| mtaylor | is there a way to tell exercise.sh to not run it? | 19:43 |
| dtroyer | yes, SKIP_EXERCISES= inlocalrc | 19:43 |
| dtroyer | I think it would be SKIP_EXERCISES=client_env | 19:44 |
| mtaylor | I'm fine with that ... jeblair, do you see any issues with doing that ^^^ | 19:45 |
| *** flaviamissi has quit IRC | 19:48 | |
| *** dolphm has joined #openstack-dev | 19:53 | |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Remove utils.runthis(). https://review.openstack.org/3587 | 19:54 |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Add initiator to initialize_connection https://review.openstack.org/3586 | 19:56 |
| *** flaviamissi has joined #openstack-dev | 19:56 | |
| openstackgerrit | Vish Ishaya proposed a change to openstack/nova: Add initiator to initialize_connection https://review.openstack.org/3586 | 19:58 |
| vishy | lots of pushes :) | 19:58 |
| *** bengrue has joined #openstack-dev | 20:02 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/openstack-planet: Added chmouel to Planet Openstack https://review.openstack.org/3588 | 20:06 |
| *** jmolinaso has quit IRC | 20:06 | |
| openstackgerrit | Josh Kearney proposed a change to openstack/python-novaclient: Set the default service name to 'Nova'. https://review.openstack.org/3589 | 20:07 |
| *** adjohn has quit IRC | 20:09 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack/openstack-planet: Add hackergotchi image for chmouel https://review.openstack.org/3590 | 20:10 |
| *** martines has quit IRC | 20:12 | |
| *** hashar has joined #openstack-dev | 20:14 | |
| *** nati2 has quit IRC | 20:15 | |
| *** sandywalsh has quit IRC | 20:16 | |
| *** martines has joined #openstack-dev | 20:17 | |
| openstackgerrit | Paul McMillan proposed a change to openstack/horizon: Converted Quantum to Datatables. https://review.openstack.org/3336 | 20:18 |
| openstackgerrit | Verification of a change to openstack/python-novaclient failed: Set the default service name to 'Nova'. https://review.openstack.org/3589 | 20:22 |
| *** novas0x2a|laptop has quit IRC | 20:23 | |
| *** novas0x2a|laptop has joined #openstack-dev | 20:24 | |
| *** anotherjesse has quit IRC | 20:24 | |
| *** davlap has quit IRC | 20:25 | |
| *** dprince has quit IRC | 20:27 | |
| *** davlap has joined #openstack-dev | 20:29 | |
| jk0 | mtaylor: you around to look at https://jenkins.openstack.org/job/gate-integration-tests-devstack-vm/955/console ? I believe a NOVA_REGION_NAME env var needs to be set | 20:30 |
| mtaylor | jk0: yup. | 20:32 |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Use os.path.basename() instead of string splitting. https://review.openstack.org/3591 | 20:32 |
| jk0 | mtaylor: thanks :) | 20:32 |
| mtaylor | jk0: any idea to what it should be set? and/or is the config change correct/expected? | 20:33 |
| jk0 | the defaults should just work.. I'm just guessing with that | 20:34 |
| jk0 | hm | 20:34 |
| jk0 | not much to go on | 20:34 |
| mtaylor | hrm indeed | 20:35 |
| *** hashar has quit IRC | 20:37 | |
| smoser | vishy, remembering something above | 20:37 |
| smoser | i think you already had 'dd' in a list of "things we sudo to do" | 20:38 |
| smoser | dd if=/dev/stdin of=/dev/stdout ~= cat , right? | 20:38 |
| jeblair | dtroyer, mtaylor: skipping that tests sounds fine to me | 20:38 |
| jeblair | dtroyer: you can patch openstack-ci/slave_scripts/devstack-vm-gate-host.sh to change the localrc that the devstack test uses | 20:39 |
| dtroyer | jeblair: cool, will do | 20:40 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/nova: Allow for auditing of API calls. https://review.openstack.org/3592 | 20:40 |
| annegentle | jeblair: or mtaylor: if I want to run a tool called pandoc on the Jenkins server, how do I get started? It's to take markdown to Docbook, then run the maven job. | 20:42 |
| mtaylor | annegentle: first step would be to get pandoc installed ... then we'd just need to change the doc build script | 20:43 |
| jeblair | annegentle: i love pandoc. :) | 20:43 |
| annegentle | woo | 20:43 |
| annegentle | ok, I'm looking at the Jenkins job config, and can I just add a pre-build step? | 20:43 |
| jeblair | mtaylor: it's haskell | 20:43 |
| *** bepernoot has quit IRC | 20:43 | |
| *** dolphm has quit IRC | 20:43 | |
| annegentle | do you want me to put this request in through here or is it a CI request o? | 20:44 |
| annegentle | er, or is there a process I should follow? | 20:44 |
| *** anotherjesse has joined #openstack-dev | 20:44 | |
| annegentle | I've tested it locally, not sure what else you'd like me to ensure prior to installing on Jenkins | 20:44 |
| openstackgerrit | Dean Troyer proposed a change to openstack/openstack-ci: Add client-env to SKIP_EXERCISES https://review.openstack.org/3593 | 20:44 |
| jeblair | 1.8.1.1-1ubuntu2 is available in ubuntu | 20:44 |
| jeblair | annegentle: will that version be okay? | 20:44 |
| annegentle | ayup, I've got pandoc 1.8.1.1 locally | 20:45 |
| jeblair | annegentle: and you can file a bug against the openstack-ci project for things like this: https://bugs.launchpad.net/openstack-ci | 20:45 |
| *** dolphm has joined #openstack-dev | 20:45 | |
| annegentle | jeblair: ok, logging | 20:46 |
| annegentle | jeblair: thanks guys, what's your guess for when this can get done? | 20:48 |
| annegentle | jeblair: the new markdown doc is approved, just need to test the automation layer to publish | 20:48 |
| *** lloydde has joined #openstack-dev | 20:48 | |
| mtaylor | annegentle: should probably be able to get to it this week | 20:52 |
| annegentle | mtaylor: great, thanks a lot | 20:52 |
| *** dwalleck has quit IRC | 20:52 | |
| annegentle | mtaylor: a LOT | 20:52 |
| annegentle | :) | 20:52 |
| *** dwalleck has joined #openstack-dev | 20:54 | |
| dtroyer | jeblair: I pushed https://review.openstack.org/3593 to skip the new devstack exercise | 20:54 |
| *** sandywalsh has joined #openstack-dev | 20:55 | |
| jeblair | dtroyer: approved, jenkins should be testing that change now (it's gated along with the other projects and devstack) | 20:56 |
| *** bhall has joined #openstack-dev | 21:01 | |
| *** bhall has quit IRC | 21:01 | |
| *** bhall has joined #openstack-dev | 21:01 | |
| openstackgerrit | Monty Taylor proposed a change to openstack/openstack-ci: Update launch_slave to use paramiko. https://review.openstack.org/3563 | 21:02 |
| *** dubsquared has quit IRC | 21:02 | |
| mtaylor | jeblair: updated ^^ | 21:02 |
| openstackgerrit | Verification of a change to openstack/python-novaclient failed: Set the default service name to 'Nova'. https://review.openstack.org/3589 | 21:03 |
| *** stuntmac_ has quit IRC | 21:04 | |
| termie | ayoung: more around now | 21:05 |
| ayoung | termie, I'm looking into the LDAP backend for KSredux | 21:05 |
| ayoung | BTW, keystone segment of the meeting is happening in #openstack-meeting right now | 21:05 |
| *** bpg has quit IRC | 21:07 | |
| *** dwalleck has quit IRC | 21:07 | |
| *** stuntmachine has joined #openstack-dev | 21:07 | |
| *** stuntmachine has quit IRC | 21:07 | |
| *** stuntmachine has joined #openstack-dev | 21:07 | |
| openstackgerrit | Verification of a change to openstack/nova failed: Use os.path.basename() instead of string splitting. https://review.openstack.org/3591 | 21:08 |
| vishy | ayoung: there is a pretty good ldap fake in nova which could be very helpful | 21:08 |
| ayoung | vishy, yes, I'm using that as the starting point | 21:09 |
| vishy | also the ldap in nova was rewritten as the ldap in keystone heavy too | 21:09 |
| openstackgerrit | Monty Taylor proposed a change to openstack/openstack-ci: Update launch_slave to use paramiko. https://review.openstack.org/3563 | 21:09 |
| vishy | so there are a couple of references there :) | 21:09 |
| vishy | I don't claim to be good enough at ldap to claim that I did it the right way though :) | 21:09 |
| ayoung | vishy, thanks. We should chat after this meeting, just so I can get a brain dump from you on the LDAP approach | 21:10 |
| *** berendt has quit IRC | 21:10 | |
| *** dwalleck has joined #openstack-dev | 21:11 | |
| *** flaviamissi has quit IRC | 21:12 | |
| vishy | ayoung: you assume i remember :) | 21:12 |
| ayoung | vishy, nah, I just assume you can fake it. | 21:12 |
| *** anotherjesse1 has joined #openstack-dev | 21:13 | |
| *** troytoman-away is now known as troytoman | 21:13 | |
| *** anotherjesse has quit IRC | 21:13 | |
| *** stuntmachine has quit IRC | 21:14 | |
| *** dwalleck has quit IRC | 21:16 | |
| *** dwalleck has joined #openstack-dev | 21:16 | |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Fix broken devref docs. https://review.openstack.org/3594 | 21:17 |
| *** adjohn has joined #openstack-dev | 21:18 | |
| *** stuntmachine has joined #openstack-dev | 21:18 | |
| *** lts has quit IRC | 21:20 | |
| *** martines has quit IRC | 21:20 | |
| *** joesavak has quit IRC | 21:22 | |
| *** anotherjesse1 has quit IRC | 21:22 | |
| openstackgerrit | Monsyne Dragon proposed a change to openstack/nova: Make parsing of usage stats from XS more robust. https://review.openstack.org/3595 | 21:23 |
| *** martines has joined #openstack-dev | 21:24 | |
| *** pixelbeat has joined #openstack-dev | 21:27 | |
| termie | zns: who is using keystone (the python one)? | 21:32 |
| *** andrewsmedina has quit IRC | 21:33 | |
| *** hashar has joined #openstack-dev | 21:33 | |
| *** dtroyer has quit IRC | 21:39 | |
| *** dtroyer has joined #openstack-dev | 21:39 | |
| *** hashar has quit IRC | 21:41 | |
| *** markwash has joined #openstack-dev | 21:44 | |
| *** jdg has joined #openstack-dev | 21:45 | |
| openstackjenkins | Project nova-docs build #2077: SUCCESS in 3 min 1 sec: https://jenkins.openstack.org/job/nova-docs/2077/ | 21:48 |
| openstackjenkins | * rbryant: Remove utils.runthis(). | 21:48 |
| openstackjenkins | * rbryant: Use os.path.basename() instead of string splitting. | 21:48 |
| *** stuntmachine has quit IRC | 21:49 | |
| *** stuntmachine has joined #openstack-dev | 21:50 | |
| ayoung | termie, is the goal that keystone-manage goes away and we use the 'keystone' command to do everything that keystone manage does? | 21:56 |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Use "display_name" in "nova-manage vm list". https://review.openstack.org/3596 | 21:56 |
| termie | ayoung: keystone-manage still exists and is the cli | 21:56 |
| termie | ayoung: keystone is the service itself | 21:56 |
| termie | keystone-manage does one thing (db_sync) for setup, and the rest is just a wrapper around keystoneclient | 21:57 |
| ayoung | termie, ok, so it looks like the nova approach is to do everything via the remote API. Is keystone going to try and do the same? | 21:57 |
| ayoung | keystone-client | 21:57 |
| termie | ayoung: keystoneclient is python-keystoneclient, the library most projects use to interact with keystone | 21:57 |
| termie | ayoung: everything is handled via the api (and admin api) except for the db_sync operation | 21:58 |
| termie | which is the migration | 21:58 |
| termie | and initial db setup | 21:58 |
| termie | ./bin/keystone-manage without any args should give you help output | 21:58 |
| *** hashar has joined #openstack-dev | 21:59 | |
| ayoung | termie, right...I was just learning the pre KSL keystone-manage...threw me for a loop | 21:59 |
| zns | termie: I don't have a comprehensive list. I know Rackspace does. | 21:59 |
| *** hashar has quit IRC | 21:59 | |
| termie | ayoung: yeah, the ksl keystone-manage is a bit different, the two projects have a different data model and a different way of interacting with the serivce | 22:00 |
| termie | ayoung: pre-ksl it was doing direct db ops | 22:00 |
| termie | ayoung: so you didn't even have to have keystone running | 22:00 |
| termie | zns: rackspace is using the python version or my team at rackspace is? | 22:01 |
| *** dwalleck has quit IRC | 22:01 | |
| termie | zns: do you think you can email me a list of those that you are aware of? | 22:01 |
| termie | zns: i need to start talking to them | 22:01 |
| *** negronjl has quit IRC | 22:01 | |
| *** negronjl has joined #openstack-dev | 22:01 | |
| termie | zns: my understanding was that rackspace has a java implementation as well | 22:01 |
| *** Gordonz has quit IRC | 22:02 | |
| *** pmyers has quit IRC | 22:02 | |
| termie | zns: or one of something similar | 22:02 |
| *** LinuxJedi has quit IRC | 22:02 | |
| *** stuntmachine has quit IRC | 22:02 | |
| Daviey | vishy: it seems to me, that vendors will still land crack at the last minute.. but just in a different tree :/ | 22:02 |
| Ryan_Lane | vishy: right now it's possible to do some extension via drivers, and api extensions, but there is still a need to directly modify core for many features | 22:02 |
| Daviey | which IMO, is no different to using the stable/ process | 22:02 |
| Ryan_Lane | even features that should never appear in core | 22:02 |
| *** LinuxJedi has joined #openstack-dev | 22:02 | |
| vishy | Daviey: true :) | 22:03 |
| vishy | Ryan_Lane: I think the goal with those features is to find the right abstraction | 22:03 |
| Daviey | vishy: it feels to me, that there should be a deadline for non-release blocker fixes. | 22:03 |
| Daviey | and a broken driver isn't that | 22:03 |
| vishy | Ryan_Lane: so that core can be extended | 22:03 |
| *** jdg has quit IRC | 22:03 | |
| Daviey | If they miss the deadline, they have to go the stab;e/ root. | 22:03 |
| openstackgerrit | Mark Washenberger proposed a change to openstack/nova: Add mkswap to rootwrap https://review.openstack.org/3597 | 22:03 |
| Ryan_Lane | well, why not some form of hook system, or some other normal form of extensibility? | 22:04 |
| *** dolphm has quit IRC | 22:04 | |
| openstackgerrit | Brian Waldon proposed a change to openstack/nova: Use single call in ExtendedStatus extension https://review.openstack.org/3598 | 22:04 |
| vishy | Ryan_Lane: right, if there are more hooks needed, we should add them | 22:04 |
| Ryan_Lane | there are hooks right now? | 22:04 |
| vishy | Ryan_Lane: I don't think it is easy to see where those hooks are until you have something that you want to add that needs one | 22:04 |
| Ryan_Lane | I totally agree. hooks should occur naturally over time | 22:05 |
| vishy | Ryan_Lane: well in the general sense of the term, yes | 22:05 |
| Ryan_Lane | I didn't realize we already had a hooks system | 22:05 |
| vishy | Ryan_Lane: although I don't know if you are talking about something specific | 22:05 |
| Daviey | vishy: If this goes ahead, it makes it easier for me (who doesn't care about driver/foobar) to break it (and not notice or care), and just expect foobar carers to fix it | 22:05 |
| Daviey | ATM, i can't break foobar, or if i do - i need to work with foobar to fix it, before my code lands. | 22:05 |
| Ryan_Lane | are you talking about drivers? | 22:05 |
| vishy | Daviey: that is a good point, but on the other hand it means that we have lots of code being proposed for drivers | 22:05 |
| vishy | which none of us has a clue about | 22:05 |
| vishy | so we just go, if you say so and click merge | 22:06 |
| Daviey | heh, if the tests pass, ship it! :) | 22:06 |
| vishy | Daviey: maybe the right way to do it with some kind of subpackage system | 22:06 |
| Ryan_Lane | +1 on subpackage | 22:06 |
| vishy | so they can work in their own branch and we could pull from their stuff every so often | 22:06 |
| vishy | so they have a version of the code in core that we test against | 22:06 |
| Daviey | That is like feature trees, right? | 22:07 |
| *** stuntmachine has joined #openstack-dev | 22:07 | |
| *** stuntmachine has joined #openstack-dev | 22:07 | |
| vishy | Daviey: it means that they can still ship newer versions of their code if they want | 22:07 |
| Daviey | So, this is more about devolving ownership of things -core doesn't follow as closely? | 22:07 |
| vishy | so we would have 1.0 in, but they could release 1.1 or 1.2 separately | 22:07 |
| Daviey | vishy: This makes life harder for distro's :) | 22:07 |
| vishy | Daviey: yeah, all of the vendors are scrambling like mad to get their stuff into essex | 22:08 |
| Daviey | having promised working versions.. is a nice feature. | 22:08 |
| vishy | Daviey: they could provide their own ppas | 22:08 |
| *** rnirmal has quit IRC | 22:08 | |
| vishy | which a user can choose to use | 22:08 |
| vishy | Daviey: I'm just brainstorming here btw | 22:08 |
| Daviey | if it turns into one big driver tree, distro might have to decide between superfast foo driver or optimal bar? | 22:08 |
| *** stuntmachine has quit IRC | 22:08 | |
| vishy | Daviey: I'm thinking more like: | 22:09 |
| Daviey | vishy: right.. and i'm just stomping on everything, sorry | 22:09 |
| vishy | if a consumer decides that they really want to use a solidfire san | 22:09 |
| vishy | then I don't think they mind using solidfire's special ppa for up-to the minute drivers | 22:09 |
| vishy | and they should have an easy way to get those | 22:09 |
| mtaylor | hey all - we're moving paste.openstack.org to a new server - it's possible over the next 5 minutes or so that you could lose a paste | 22:09 |
| vishy | even though they don't matter to 99% of nova users | 22:10 |
| vishy | mtaylor: are you moving it to hastie.com ? | 22:10 |
| Daviey | vishy: I really think there needs to be more love for the stable tree :) | 22:10 |
| Daviey | regular releases from the stable tree will help this. | 22:10 |
| vishy | Daviey: we all love the stable tree | 22:10 |
| *** jdg has joined #openstack-dev | 22:10 | |
| Daviey | .. and, all distro users benefit. | 22:11 |
| mtaylor | vishy: nah - just another cloud server - but this time one that's actually properly configuration managed and what not. you know - init scripts | 22:11 |
| Daviey | vishy: it is worth remembering that once a user goes custom ppa, they are bound to that ppa... no promise of upgrade path. | 22:12 |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Ties quantum, melange, and nova network model https://review.openstack.org/3309 | 22:14 |
| Daviey | vishy: I do agree that there will be a time, when nova-core cannot be expected to understand everyfink.. Following the kernel dev process, having seperate working groups does make sense. | 22:14 |
| Daviey | altough, i refuse to call the main tree vishystack btw. | 22:14 |
| *** _adjohn has joined #openstack-dev | 22:14 | |
| *** dolphm has joined #openstack-dev | 22:14 | |
| * mtaylor will now only call the main tree vishystack | 22:15 | |
| *** adjohn has quit IRC | 22:16 | |
| *** _adjohn is now known as adjohn | 22:16 | |
| * vishy votes for mtaylor | 22:16 | |
| vishy | :) | 22:16 |
| deshantm | vishy: is your idea also to potentially have core plugins? or does that not even make sense? | 22:16 |
| vishy | deshantm: in my brainstorm, i think core plugins would make sense | 22:17 |
| vishy | deshantm: they would be the ones that we test against | 22:17 |
| deshantm | right, but optional for deployers | 22:17 |
| vishy | deshantm: yes, we already have the equivalent in the various drivers that we have | 22:18 |
| vishy | deshantm: we just ship all of the optional components in trunk right now | 22:18 |
| vishy | deshantm: I'm trying to decrease the cognitive load of nova a bit | 22:18 |
| vishy | nova: you're going on a diet! | 22:18 |
| deshantm | vishy: from 10,000 feet it seems like a great idea | 22:18 |
| vishy | :) | 22:18 |
| vishy | Daviey makes some great points about it complicating things for packagers though | 22:19 |
| deshantm | maybe more complicated, but probably still better | 22:19 |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Ties quantum, melange, and nova network model https://review.openstack.org/3309 | 22:21 |
| deshantm | vishy: when you say we already have this model for drivers, do you mean that you can plug the libvirt or XenAPI driver for example? | 22:21 |
| vishy | deshantm: yes, the flags --compute_manager or --compute_driver can be changed to use other managers and drivers | 22:24 |
| *** martine has quit IRC | 22:24 | |
| vishy | deshantm: it just has to point to an importable python callable, so the code can be shipped separately | 22:24 |
| deshantm | vishy: neat | 22:25 |
| openstackgerrit | anotherjesse proposed a change to openstack/nova: remove non-supported ec2 admin extensions https://review.openstack.org/3599 | 22:30 |
| *** apevec has quit IRC | 22:31 | |
| *** andrewsmedina has joined #openstack-dev | 22:36 | |
| *** anotherjesse has joined #openstack-dev | 22:37 | |
| *** sleepsonthefloo has quit IRC | 22:39 | |
| *** anotherjesse has quit IRC | 22:44 | |
| rmk | so ksl huh | 22:46 |
| *** sandywalsh has quit IRC | 22:46 | |
| *** jdg_ has joined #openstack-dev | 22:50 | |
| termie | ayoung: heya, what do you work that you are using keystone? | 22:51 |
| termie | s/work/work on/ | 22:51 |
| ayoung | termie, now--days? Openstack | 22:51 |
| ayoung | I was on FreeIPA as of a few months back | 22:52 |
| *** markvoelker has quit IRC | 22:52 | |
| *** Remco_ has joined #openstack-dev | 22:52 | |
| termie | ayoung: are you at racksapce? sorry i don't know you i am just trying to get a list of who is trying to use it now so i can check requirements against realit | 22:53 |
| termie | y | 22:53 |
| ayoung | termie, I am at Red Hat | 22:53 |
| termie | gyee: and you are with HP? | 22:53 |
| openstackgerrit | Russell Bryant proposed a change to openstack/nova: Fix VPN ping packet length. https://review.openstack.org/3600 | 22:53 |
| termie | ayoung: coolio, thanks | 22:53 |
| gyee | termie, yes | 22:53 |
| termie | want to private message me your email addresses? | 22:53 |
| termie | oh, i have yours gyee | 22:54 |
| gyee | btw, has anybody tried running the tests for keystone redux | 22:54 |
| termie | yes | 22:54 |
| gyee | its taking forever | 22:54 |
| termie | the first run does a few checkouts | 22:54 |
| gyee | CliMasterTestCase been running for the last hour | 22:54 |
| gyee | and all failing | 22:54 |
| termie | what? weird | 22:54 |
| openstackgerrit | Christian S. Perone proposed a change to openstack/python-novaclient: Fixing a problem where a exception is silenced. https://review.openstack.org/3601 | 22:54 |
| *** cp16net has quit IRC | 22:56 | |
| *** vincentricci has joined #openstack-dev | 22:56 | |
| termie | gyee: i am not sure what that would be about unless your internet connection has failed or soemthing strange, they all run in about 10 seconds for me | 22:56 |
| termie | i'll try clearing my venv and seeing whether anything new happens | 22:57 |
| gyee | weird, I have no problem with the essex branch | 22:58 |
| gyee | so shouldn't be a network issue | 22:58 |
| termie | gyee: what branch are you running? | 22:58 |
| gyee | redux | 22:58 |
| termie | on which repo | 22:58 |
| gyee | keystone | 22:58 |
| termie | k, that is a weird merge branch | 22:59 |
| termie | so i haven't run anything against that | 22:59 |
| termie | main dev has been occuring at https://github.com/termie/keystonelight for some time, that branch was created last night | 22:59 |
| termie | i'll go look at it now | 22:59 |
| *** dolphm has quit IRC | 22:59 | |
| gyee | k thanks | 22:59 |
| *** sandywalsh has joined #openstack-dev | 23:00 | |
| *** vincentricci has quit IRC | 23:00 | |
| *** vincentricci has joined #openstack-dev | 23:00 | |
| *** rkukura has quit IRC | 23:01 | |
| gyee | uh oh, my firewall is block 207.97.227.239 | 23:01 |
| gyee | blocking | 23:02 |
| jeblair | termie: the redux _branch_ is exactly your branch. there is a merge commit that merges it into master (and that's certainly weird) but the branch itself is straight-up. | 23:02 |
| *** maoy has joined #openstack-dev | 23:02 | |
| termie | jeblair: hmm | 23:02 |
| gyee | why the tests are talking to github.com over https? | 23:02 |
| bcwaldon | comstud: are you around? | 23:03 |
| maoy | hi all.. | 23:03 |
| termie | gyee: github serves requests over https | 23:03 |
| mtaylor | gyee: I've got a patch to change that - I need to rebase it aginst current work though and then argue with termie about it some | 23:03 |
| gyee | I mean essex tests doesn't talk to github.com at all | 23:03 |
| maoy | a quick question about rpc.call: if the topic is "network" and there are multiple nodes running the network service, how do I know which node will receive the rpc call? Is it random? Thanks. | 23:04 |
| openstackjenkins | Project nova-docs build #2078: SUCCESS in 3 min 42 sec: https://jenkins.openstack.org/job/nova-docs/2078/ | 23:04 |
| openstackjenkins | Brian Waldon: Use single call in ExtendedStatus extension | 23:04 |
| termie | gyee: is talking to github a problem for your firewall? | 23:04 |
| mtaylor | termie: the reason I was trying to change that is that we run tests frequently enough in openstack's jenkins that we see an unfortunately high amount of service failure from github | 23:04 |
| *** ayoung has quit IRC | 23:04 | |
| gyee | termie, over https yes, http no | 23:04 |
| termie | mtaylor: this isn't a service failure from github that gyee is running into | 23:04 |
| mtaylor | termie: I know - I was just using it as a good excuse to start the convo with you about the patch I need landed ;) | 23:05 |
| termie | mtaylor: i gave you comments on your patch | 23:05 |
| gyee | my corporate firewall is too good and I am too tired to pick a fight with IT :) | 23:05 |
| mtaylor | termie: I know - I'm responding to them right now | 23:05 |
| mtaylor | termie: we try our best to not ever attempt to talk to network resources outside of our infrastructure during a test run itself, because they cause an annoyingly high amount of false negatives | 23:06 |
| termie | gyee: well, you've got a couple options, you can grep for https and change it in the three places in the code | 23:06 |
| termie | mtaylor: take it up in the code review please? | 23:06 |
| mtaylor | termie: ok | 23:06 |
| termie | gyee: you can do that then propose a change | 23:07 |
| termie | gyee: you can shell out to a computer outside the firewall | 23:07 |
| gyee | yeah, making it configurable would be nice | 23:07 |
| termie | gyee: you can propose a patch that makes it configurable | 23:07 |
| jeblair | mtaylor: i don't see a change in gerrit? | 23:07 |
| gyee | k | 23:07 |
| termie | jeblair: it is a github pull request | 23:07 |
| mtaylor | jeblair: I haven't submitted it to gerrit yet | 23:08 |
| termie | jeblair: but it should 3-4 different patches | 23:08 |
| mtaylor | jeblair: I sent it in a couple of weeks ago ... working on rebasing right now | 23:08 |
| *** flaviamissi has joined #openstack-dev | 23:08 | |
| jeblair | oh ok. should probably go into gerrit. i'm not planning on pulling/pushing that tree again | 23:08 |
| mtaylor | yes | 23:08 |
| termie | jeblair: how do we get our work in the ksl tree into redux? | 23:09 |
| termie | jeblair: i saw that you said git review redux | 23:09 |
| termie | can i just do that periodically from master? | 23:09 |
| jeblair | you can do your work in the redux branch and review/merge it with gerrit, just like any other branch | 23:10 |
| jeblair | (stable/diablo, etc) | 23:10 |
| termie | so i'll checkout a copy of the redux branch and rebase master changes on to it, then `git review redux` | 23:10 |
| termie | does that sound viable to you? | 23:11 |
| *** maoy has quit IRC | 23:11 | |
| termie | rebase/cherry pick | 23:11 |
| jeblair | yes, that should work | 23:11 |
| jeblair | the first change to go in may need to make the tests work. | 23:13 |
| termie | s/first change/changes/ s/may/will/ ? | 23:13 |
| termie | why would the "first change" be special? | 23:13 |
| mtaylor | because no changes will land if they don't pass tests | 23:14 |
| termie | the redux branch is a branch | 23:14 |
| mtaylor | the redux branch runs jenkins jobs just like any other branch | 23:14 |
| termie | the merge prop is going to be a wallop | 23:14 |
| termie | i am not submitting changes to the redux branch, we are submitting the redux branch to master | 23:14 |
| jeblair | the merge will have to pass tests as well | 23:14 |
| termie | so the merge prop needs to pass tests | 23:14 |
| termie | the changes to redux branch to get tests passing should not | 23:15 |
| mtaylor | so - folks communicated that redux was ready to be pushed in to gerrit so that subsequent work by folks could go through code review/jenkins as normal ... | 23:15 |
| mtaylor | perhaps putting the redux branch in gerrit right now was slightly early? | 23:16 |
| termie | which folks communicated that? | 23:16 |
| jeblair | if it can't pass tests, or will need a lot of work to pass tests, it sounds like that was premature. | 23:16 |
| termie | my understanding is that we have a merge prop that needs to go in, once that is in works goes through the gerrit workflow | 23:16 |
| gyee | termie, turned out I had to change KEYSTONECLIENT_REPO to 'http://...' | 23:16 |
| termie | gyee: rather than git:// | 23:17 |
| gyee | yes | 23:17 |
| gyee | firewall is blocking git:// | 23:17 |
| gyee | I know, stupid | 23:17 |
| gyee | but now I see 30 failed tests | 23:17 |
| gyee | 8 skipped, 30 errors | 23:17 |
| termie | gyee: well that is still too many so you may want to check the other urls | 23:18 |
| mtaylor | it was in our email chain with jesse - but as can happen with email chains, it sounds like we may have missed each other | 23:18 |
| termie | gyee: there are 6 errors right now because keystoneclient just changed its api | 23:18 |
| jeblair | it's also in http://etherpad.openstack.org/essex-keystone-proposal | 23:18 |
| gyee | AttributeError: 'RoleManager' object has no attribute 'add_user_to_tenant' | 23:18 |
| gyee | doesn't seem to do with url | 23:18 |
| termie | gyee: that is the keystoneclient error | 23:19 |
| termie | there are 6 of those | 23:19 |
| termie | not 30 | 23:19 |
| termie | so if you ahve 30, you probably are running into something else | 23:19 |
| termie | there are three places code is being checked out | 23:19 |
| termie | you can also grep -R tests _REPO | 23:19 |
| termie | erm switch arg order | 23:19 |
| gyee | No module named cli.app | 23:20 |
| termie | gyee: did you use the virtualenv? | 23:20 |
| gyee | no | 23:20 |
| termie | gyee: have you installed the dependencies? | 23:20 |
| termie | you have to do one or ther other | 23:20 |
| termie | tools/pip-requires and tools/pip-requires-test | 23:20 |
| gyee | thought I did | 23:20 |
| gyee | but lemme do it once more | 23:21 |
| mtaylor | termie: I'll send an email. but I think ksl should continue dev on github until we're actually at the point where we want to do the merge/start having jenkins run on it, yeah? | 23:21 |
| termie | i have also heard that pycli might have turned into pyCLI but, the pip requires should ahve the proper one | 23:21 |
| termie | mtaylor, jeblair: i don't see what we are talkinga bout in this etherpad, but the only thing that makes any sense is my mind is for the merge prop to be the thing gated on tests | 23:21 |
| mtaylor | termie: ok. I think we agree with each other | 23:22 |
| termie | mtaylor, jeblair: and for the redux branch that the merge prop is based on to be a branch | 23:22 |
| termie | mtaylor, jeblair: we are not submitting merge props to that branch, we are getting the branch ready for the merge prop | 23:22 |
| jeblair | it is a branch, i think the information that's missing is that we test all branches | 23:22 |
| *** sleepsonthefloo has joined #openstack-dev | 23:22 | |
| termie | mtaylor, jeblair: getting it in to the gerrit system is to identify the problems that arise from gerrit | 23:22 |
| jeblair | any change to a branch in gerrit gets reviewed, must be approved, and must pass tests | 23:22 |
| jeblair | if it's not ready for that, you should keep doing development outside of gerrit | 23:23 |
| termie | mtaylor, jeblair: we are doing the development outside of gerrit, but we need the gerrit tests to pass | 23:23 |
| termie | mtaylor, jeblair: at the moment you say there are failing tests in gerrit so we need the branch to be able to identify those | 23:24 |
| mtaylor | actually - I think there's another way here | 23:24 |
| termie | mtaylor, jeblair: but requiring the "first change" to fix all tests means some ugly squashed commit that touches too many things | 23:24 |
| mtaylor | which is that we copy the jenkins jobs that we would be running on keystone, and we point it to poll the github ksl repo | 23:25 |
| *** bsza has quit IRC | 23:25 | |
| mtaylor | ... so that you can get the feedback on the branch, and when it's green, we know we're good - I think the gerrit part of this is the part that's confusing conversation | 23:25 |
| termie | that's like saying the obstacle is what's in the way here | 23:26 |
| jeblair | mtaylor: why is that necessary? why not just say "these are the commands jenkins runs, these need to work before we're ready to merge?" | 23:26 |
| termie | i don't really care anymore and would be quite happy to have that list of commands from jeblair | 23:27 |
| mtaylor | that is also entirely possible | 23:27 |
| termie | so i can fix this crap right now and be through with it | 23:27 |
| mtaylor | great. that sounds easy | 23:27 |
| *** flaviamissi has quit IRC | 23:27 | |
| termie | gyee: did you go cahnge the other git:// urls and try running the rests again? | 23:28 |
| termie | gyee: you'll still have the 6 errors, though i expect to fix them shortly | 23:28 |
| *** flaviamissi has joined #openstack-dev | 23:28 | |
| *** flaviamissi has quit IRC | 23:33 | |
| termie | jeblair: are you emailing that stuff at me? | 23:33 |
| jeblair | mtaylor is | 23:33 |
| *** flaviamissi has joined #openstack-dev | 23:34 | |
| *** anotherjesse has joined #openstack-dev | 23:34 | |
| vishy | maoy: it is random | 23:36 |
| *** flaviamissi has quit IRC | 23:38 | |
| *** zzed has quit IRC | 23:41 | |
| *** _adjohn has joined #openstack-dev | 23:41 | |
| *** vincentricci_ has joined #openstack-dev | 23:41 | |
| *** vincentricci has quit IRC | 23:44 | |
| *** vincentricci_ is now known as vincentricci | 23:44 | |
| *** adjohn has quit IRC | 23:44 | |
| *** _adjohn is now known as adjohn | 23:44 | |
| openstackgerrit | Trey Morris proposed a change to openstack/nova: Ties quantum, melange, and nova network model https://review.openstack.org/3309 | 23:50 |
| andrewbogott | If I implement a volume driver without snapshot support, what will break? Just snapshots? Or will other things (e.g. host migration) fall apart as well? | 23:51 |
| *** PotHix has quit IRC | 23:53 | |
| *** mattray has quit IRC | 23:54 | |
| *** anotherjesse_ has joined #openstack-dev | 23:54 | |
| *** anotherjesse has quit IRC | 23:56 | |
| *** russellb has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!